Should You Capture Analog Console Output Directly or Capture the Scaler HDMI Output?

You are splitting your console's analogue output to a lag-free display and a capture card, and you have two paths: feed the raw signal into a digitizer like a RetroTINK-2X Pro in passthrough mode, or let a processor like the OSSC or RetroTINK-5X Pro handle the image first and capture its HDMI output. The decision to capture directly versus through a scaler comes down to one thing: who controls the analogue-to-digital conversion. The direct path puts the capture card's ADC in charge. The scaler path puts a dedicated, tested ADC in charge. The rest follows from that.

In the direct path, your signal chain is console analogue output, RGB or component cable, a scaler set to passthrough mode, HDMI into the capture card. The RetroTINK-2X Pro in passthrough mode adds zero frames of latency. The OSSC in passthrough mode adds less than one scanline. The capture card itself handles the conversion from analogue to digital.

The problem is that most capture cards are not built for 240p game signals. A typical Elgato or AVerMedia card expects a standard video resolution and treats 240p as 480i. That triggers a destructive deinterlace that halves vertical resolution and adds combing artifacts. Even if the card locks to the signal, its ADC bit depth is often lower than what a dedicated scaler provides. The result is crushed blacks and clipped whites on component sources. The digital conversion loses information at the extremes of the voltage range, and that loss is permanent.

This path works well for systems that output 480i or higher natively, like a PlayStation 2 over component. For 240p consoles, the capture card's misinterpretation of the signal is a failure mode you cannot fix in software.

In the scaler path, the signal chain is console analogue output, RGB or component cable, a processor with active scaling, HDMI into the capture card. The scaler digitizes the analogue signal with its own ADC, applies line multiplication or framebuffer scaling, and outputs a clean HDMI signal that the capture card treats as a standard video source.

The RetroTINK-5X Pro outputs 1080p or 1440p with motion-adaptive deinterlacing for 480i sources. The OSSC outputs resolutions from 480p to 1600x1200, depending on the line multiplication mode. The RetroTINK-4K goes up to 4K with custom modelines. The capture card sees a standard HDMI signal and has no reason to misinterpret it.

The trade-off is latency. The RetroTINK-5X Pro adds less than one frame in framelock mode. The OSSC adds less than one scanline in pure line multiplier mode. For most games, this is imperceptible. For rhythm games or competitive shmups, the direct path still wins on lag.

The single biggest difference between the two paths is analogue-to-digital sampling quality. A capture card's ADC is a general-purpose component designed to handle a range of video standards. A scaler's ADC is a specialised component chosen for its performance with game console signals.

The RetroTINK-5X Pro uses a high-bit-depth ADC that preserves the full voltage range of the analogue signal. The OSSC uses a similar approach. Direct capture through a typical Elgato or AVerMedia card loses detail in the shadows and highlights. The crushed blacks and clipped whites on a direct component capture are a direct result of the capture card's ADC bit depth being insufficient for the dynamic range of a game console's analogue output.

For RGB SCART sources, the difference is visible on any monitor that can show a full greyscale ramp. The direct path shows banding in the darkest and lightest steps. The scaler path shows a clean, continuous gradient. This is not a subtle difference. It is the difference between a recording that looks like the original and one that looks like a compressed version of it.

For 480i sources like PlayStation 2 games, deinterlacing quality determines whether the recording looks smooth or flickers. The direct path hands the field processing to the capture card. Most capture cards use a simple bob or weave method that either blurs motion or produces combing artifacts.

The scaler path lets you control the process. The RetroTINK-5X Pro uses motion-adaptive deinterlacing that reconstructs full vertical resolution from alternating fields. The GBS-Control also offers motion-adaptive processing, though with some chroma artifacts on fast-moving red objects. The OSSC has no deinterlacer and treats 480i as a pass-through or bob-deinterlaced signal, which makes it a poor choice for 480i capture unless you use the OSSC Pro's framebuffer mode.

For a capture setup focused on PlayStation 2, Sega Dreamcast, or any console that switches between 240p and 480i, the RetroTINK-5X Pro is the correct tool. Its motion-adaptive field reconstruction eliminates flicker without introducing the latency of a full-frame buffer.

Why Handshakes Fail

The HDMI handshake between the scaler and the capture card can fail. The OSSC in DVI mode outputs a signal that some capture cards reject because the EDID data does not match what the card expects. The RetroTINK-4K's custom modelines can also trigger handshake loops, where the scaler and capture card continuously renegotiate the connection, causing a blinking black screen every few seconds.

Reliable Fixes

Set the OSSC to HDMI mode instead of DVI mode. For the RetroTINK-4K, use one of the standard output resolutions listed in the user manual rather than a custom modeline. If the capture card still fails to lock, add a cheap HDMI EDID emulator or splitter between the scaler and the capture card. The Shmups Forum consensus is that a Dr. HDMI or a similar EDID management device resolves most handshake issues without adding measurable latency.

For the RetroTINK-5X Pro at 1440p, some capture cards do not support 1440p60 input at all. Check your capture card's specifications before building a setup around this resolution. The Elgato HD60 X supports 1080p60 capture, not 1440p60. The Elgato 4K60 Pro Mk.2 and AVerMedia Live Gamer 4K support 4K60 HDR input and can handle 1440p60, but not all firmware versions work correctly.

Chroma subsampling determines whether red and blue edges on pixel art remain sharp or blur into the background. 4:4:4 preserves full colour resolution. 4:2:2 and 4:2:0 discard colour information, which smears the edges of sprites.

The direct capture path's chroma subsampling depends entirely on the capture card. Most consumer capture cards record in 4:2:0, which is fine for video but destructive for pixel art. The scaler path can output 4:4:4 from an RGB SCART source, and if the capture card accepts 4:4:4 input, the recording preserves the original pixel edges.

The Shmups Forum consensus is that the scaler path with an RGB SCART source into a RetroTINK-5X Pro or OSSC, captured on a card that supports 4:4:4 input, produces the cleanest pixel art recordings. The direct path with composite video or S-Video is never 4:4:4, because those formats already subsample chroma in the analogue domain. For composite video and S-Video, the scaler path still wins because the RetroTINK-5X Pro's comb filter separates luminance and chrominance better than a capture card's generic decoder.

Who Should Use the Scaler Path

The scaler output path suits the capture and streaming hobbyist who needs to split a signal cleanly to both a lag-free play display and a capture card without degrading either. It suits the competitive player who can tolerate sub-frame latency for correct image quality. It suits anyone recording 240p consoles for YouTube or Twitch and wants the recording to look like the game actually looked on a CRT.

Who Should Skip It

Skip the scaler path if you are a rhythm game player who perceives single-millisecond differences in input lag and must have zero added latency from any device. For that use case, direct analogue capture with a RetroTINK-2X Pro in passthrough mode is the only acceptable path. Skip it if you are recording only 480i or higher content and your capture card handles that signal correctly; the extra hardware is unnecessary. Skip it if you do not want to learn EDID handshake troubleshooting, because at some point, the scaler and capture card will refuse to talk to each other, and you will need to know how to fix it.

The single thing that most often goes wrong here is assuming any capture card can handle the OSSC's non-standard output timings. Test your specific combination of scaler, output resolution, and capture card before you build a permanent setup. The Shmups Forum has threads on every combination. Search there before you buy.