mClassic, OSSC, and RetroTINK 5X-Pro Compared on Lag, Scaling, and Compatibility

You plug a Super Famicom into a modern TV and the image is a smeared, flickering mess. That is the moment you need an upscaler. Three devices dominate the conversation: the mClassic, the OSSC, and the RetroTINK 5X-Pro. They solve the same problem with completely different architectures. Start with one question. Does your library include 480i-heavy PS2 games? The answer dictates everything.

The OSSC is a line multiplier with no framebuffer. It takes a 240p signal and repeats each scanline exactly, producing razor-sharp pixels with zero added lag measured in microseconds. It has no composite or S-Video input, and it cannot do motion-adaptive deinterlacing. The RetroTINK 5X-Pro is a framebuffer scaler. It captures a full frame, applies per-pixel motion-adaptive deinterlacing, and sends a clean 1440p signal with sub-millisecond lag. It accepts composite, S-Video, component, and RGB SCART. The mClassic is an HDMI dongle from Marseille Inc. powered by the HDMI port itself. It applies context-adaptive anti-aliasing to an already-digital 1080p signal and upscales it to 1440p. It does not accept any analogue input at all.

The architectural divide between a Marseille dongle, a line multiplier, and a framebuffer scaler is not academic. It determines what signals each device can handle, how much lag it adds, and what the final image looks like.

How The Marseille Dongle Works

The mClassic sits between your console and TV on an HDMI cable. It has no analogue inputs. It takes a 1080p digital signal, from a Nintendo Switch, a PS3, or a RetroTINK, and applies context-adaptive anti-aliasing to soften jagged edges, then scales it to 1440p. Independent Time Sleuth testing by RetroRGB shows approximately 1-2ms of processing latency, not the zero lag Marseille marketing implies. The mClassic does not generate new pixel data at 1440p; it outputs a 1440p frame containing a 1080p active image area upscaled. It cannot deinterlace 480i. It cannot generate scanlines. It will not help a console that outputs only analogue video.

How The OSSC Line Multiplier Works

The OSSC takes analogue video through a SCART or component input and multiplies each scanline. A 240p signal becomes 480p (Line2x), 720p (Line3x), or 1080p (Line5x). There is no framebuffer, so lag is measured in microseconds, the time it takes one scanline to arrive. The Shmups Forum and OSSC wiki document this as effectively zero. The downside: the OSSC does not accept composite or S-Video. It uses a bob deinterlace for 480i that flickers badly. It drops sync during SNES white flashes because its sync tolerance is tight. If your console outputs RGB SCART or component and you play only 240p games, the OSSC is the purest option.

How The RetroTINK 5X-Pro Framebuffer Works

The RetroTINK 5X-Pro captures the full frame into memory, applies motion-adaptive deinterlacing that reconstructs full resolution from 480i without flicker, and scales it to 1440p. Mike Chi designed it with a sub-millisecond framelock mode that adds 0.25ms of lag, measured by RetroRGB using a Time Sleuth. It accepts composite, S-Video, component, and RGB SCART. The frame buffer allows features the OSSC cannot match: scanline filters, HDR injection, black frame insertion, and profile saving. The cost is a more complex signal path and the risk of triple buffer mode adding 1-2 frames of lag if framelock fails.

The OSSC RetroTINK 5X input lag difference is the single most important spec for competitive players. The OSSC adds less than 0.001ms, one scanline at 15kHz. The RetroTINK 5X-Pro adds 0.25ms in framelock mode. Both are imperceptible to a human. The real difference is not the lag but what each device does to achieve it.

The OSSC achieves its lag floor by having no framebuffer. It processes each scanline as it arrives and sends it immediately. This means it cannot perform any per-frame analysis: no motion-adaptive deinterlacing, no rotation, no zoom. The RetroTINK 5X-Pro uses a small framebuffer that holds a fraction of a frame, allowing it to analyse motion across fields and reconstruct 480i without flicker. The 0.25ms penalty is the cost of that analysis. Both are far below the 10-15ms that a TV in Game Mode adds at 1080p, or the 30-40ms at 4K.

If you play rhythm games or shmups and perceive milliseconds, the OSSC is the theoretical winner. In practice, the RetroTINK 5X-Pro's 0.25ms is below the measurement threshold of most human reaction tests. The failure case is not the scaler lag. It is the TV. A modern flat-panel with Game Mode on still adds 10-15ms at 1080p. The scaler is never the bottleneck.

Attribute	mClassic	OSSC	RetroTINK 5X-Pro
Processing type	Post-process anti-aliasing	Line multiplication	Framebuffer scaling
Input signal	HDMI (digital only)	SCART, Component, VGA	Composite, S-Video, Component, SCART
Deinterlacing	None (passes 480i through)	Bob deinterlace (flickers)	Motion-adaptive (no flicker)
Measured input lag	1-2ms (Time Sleuth, RetroRGB 2019)	<0.001ms (one scanline, OSSC wiki 2018)	0.25ms framelock (Time Sleuth, RetroRGB 2021)
Max output resolution	1440p at 60Hz, 4K passthrough	1600x1200 (1600p)	2560x1440 (1440p)
Scanline filter	No	Yes (hybrid scanlines)	Yes (CRT simulate, PVM, slot mask)
Composite/S-Video input	No	No	Yes
Power source	HDMI port (bus-powered)	5V DC barrel jack	5V DC Micro-USB
Retail price (launch)	USD 99.99 (2019)	GBP 130-160 (2016)	USD 299.99 (2021)

The mClassic anti-aliasing vs OSSC sharp pixels choice is a visual philosophy. The mClassic applies context-adaptive anti-aliasing that softens pixel edges, reducing jagged lines on 3D games like those on the Nintendo Switch. The OSSC produces razor-sharp pixels with no processing, replicating the look of a CRT's electron beam hitting the phosphor. There is no right answer. Only the right answer for your game library.

The mClassic's anti-aliasing is most effective on 3D titles from the PlayStation 2, GameCube, and Switch era. It smooths the edges of polygons without blurring the entire image. On 2D pixel art from a Super NES or Mega Drive, the same algorithm softens the hard pixel edges that many retro enthusiasts want to preserve. The OSSC's line multiplication preserves every pixel exactly, making a 240p game look like a perfect grid of squares. The OSSC also supports scanline filters that simulate the gaps between CRT scanlines, which the mClassic cannot do.

The failure case: if you buy the mClassic expecting it to fix a 240p signal from an analogue console, it will not. The mClassic has no analogue input. You would need a separate analogue-to-HDMI converter first, and that converter will introduce the lag and image degradation the mClassic is meant to fix. The mClassic is for digital sources only: a Nintendo Switch, a PS3, a RetroTINK, or a modded console with an internal HDMI mod like the PS1Digital or Wii Dual.

The best scaler for PS2 480i deinterlacing is the RetroTINK 5X-Pro. The PS2 outputs the vast majority of its library at 480i, an interlaced signal that alternates fields of odd and even scanlines. A bob deinterlace, what the OSSC uses, alternates between showing only the odd lines and only the even lines, creating a visible flicker on every horizontal edge. The RetroTINK 5X-Pro's motion-adaptive deinterlacing analyses each pixel across fields and reconstructs a full-resolution progressive frame, eliminating flicker entirely.

The OSSC can pass 480i through to a TV, but the TV's own deinterlacer adds lag and often treats the signal as video, not game content. The OSSC's bob deinterlace mode is functional but produces a shimmering image that is fatiguing over a long session. The mClassic does not accept analogue 480i at all, so it is not an option for a raw PS2 signal. You would need to convert the PS2's component to HDMI first, and that converter would add 2-4 frames of lag and apply a destructive deinterlace that discards half the vertical resolution.

The RetroTINK 5X-Pro handles the PS2's 480i signal by default, detecting the resolution switch when a PS1 game transitions from a 240p menu to 480i gameplay. It locks the frame in less than a second, preventing the blank screen that cheaper scalers produce. If you own a PS2 and want to play it on a modern TV without flicker, the RetroTINK 5X-Pro is the only device in this comparison that solves the problem directly.

The OSSC has no composite or S-Video input. If your console outputs only composite, the yellow RCA jack, or S-Video, the OSSC cannot accept it. This rules out the NES, SNES (without RGB mod), N64 (without RGB mod), and many PAL consoles that lack RGB SCART output. The RetroTINK 5X-Pro accepts all three: composite, S-Video, and component (shared with composite via a single RCA jack).

Many unmodded consoles in the second-hand market output composite only. A used Sony PlayStation from a game shop will have composite and S-Video but not RGB SCART unless it is a later model or has been modded. The RetroTINK 5X-Pro's composite input cleans up the dot crawl artifacts that plague composite video, using a comb filter that separates luminance and chrominance. The OSSC cannot do this, so the only path for a composite-only console into an OSSC is through an external transcoder that converts composite to RGB SCART. That transcoder adds cost, complexity, and potential signal degradation.

The failure case: you pick up an OSSC without checking that your console outputs RGB. You plug in a composite cable, get no signal, and assume the scaler is broken. The OSSC requires RGB SCART or component input. If your console does not output either, the RetroTINK 5X-Pro is the only device in this comparison that will work without additional hardware.

The OSSC's microseconds-scale latency is the theoretical minimum. The RetroTINK 5X-Pro's 0.25ms in framelock mode is imperceptible. The mClassic's 1-2ms is still below what most players can feel. The real enemy is not the scaler but the TV. A typical flat-panel in Game Mode adds 10-15ms at 1080p and 30-40ms at 4K. That is the number you need to care about.

Measure your total system lag. Use a Time Sleuth or the 240p test suite. The 240p test suite is a homebrew ROM that outputs a flashing screen and a light sensor that measures the delay between the console's picture and the light appearing on screen. This test includes the console, the scaler, and the TV. If your total lag is above 50ms, you will feel it in a game like Super Mario World or a shmup. The scaler is rarely the source of that lag.

The failure case: you buy an OSSC to minimise lag, plug it into a TV with Game Mode off, and still feel delay. You blame the OSSC when the TV is adding 100ms of processing. Enable Game Mode on the TV. Disable all post-processing. Ensure the TV is not upscaling the scaler's output. If the OSSC outputs 1080p and the TV is a 4K panel, the TV will still scale the image, adding its own lag. A RetroTINK 5X-Pro sending 1440p reduces that TV scaling step on a 4K display.

Hong Kong runs on 220V at 50Hz. Any scaler purchased from a US or UK retailer will work with a simple plug adapter, but check the power supply. The RetroTINK 5X-Pro uses a 5V DC Micro-USB input and a 2.0A power supply; a standard phone charger works if it provides 2.0A. The OSSC uses a 5V DC barrel jack. The mClassic is bus-powered from the HDMI port and does not need a separate supply, but a low-power HDMI port on some TVs can cause signal dropouts.

The Hong Kong Retro Game Association holds meetups where you can test scalers before buying. The Retro.HK Gaming Expo at D2 Place in Lai Chi Kok, last held in 2024, has vendors demonstrating OSSC and RetroTINK units. For stock, visit 好景遊戲 (Ho King Game) in Mong Kok or 遊戲の箱庭 (Game Box) in Sham Shui Po. They carry upscalers. Bring your own cables. HDMI cables are cheap in Golden Computer Centre, but RGB SCART cables for the OSSC may need to be ordered from Insurrection Industries or HD Retrovision if you want proper shielding.

The failure case: you order an OSSC from the UK, receive a 240V UK plug, and plug it into a Hong Kong socket with a cheap travel adapter that provides 5V but insufficient current. The OSSC powers on but drops sync during gameplay. The fix is to use the correct power supply: 5V DC at 1.5A minimum for the OSSC, 2.0A for the RetroTINK 5X-Pro. Do not use a USB port on a TV to power the RetroTINK; it will not provide enough current.

The single most common failure across all three devices is a sync drop during a bright flash or a resolution switch. A SNES game that flashes white during a transition, like the title screen of Super Metroid, can cause the console's sync signal to dip in voltage. The scaler sees a loss of sync and goes black for 2-3 seconds. This happens most often on the OSSC because its sync tolerance is tight. The RetroTINK 5X-Pro is more forgiving, but can still drop sync on a poorly modded console.

Adjust the scaler's sync threshold settings. On the OSSC, enable the sync regenerator option. On the RetroTINK 5X-Pro, adjust the sync level in the input menu. If the problem persists, the cable is the likely culprit. A cheap SCART cable with thin wires cannot carry the sync voltage reliably. Use a shielded RGB SCART cable from a reputable source like RetroRGB or Insurrection Industries. The mClassic, being digital-only, does not have this problem because it never sees the raw analogue sync signal.