Experts Reveal That Most SACD Players Don’t Deliver the Pure DSD Sound Buyers Are Paying For

Even thousand-dollar players quietly switch your SACDs to PCM before you ever hear them.

Most SACD players, including models costing thousands of dollars, convert DSD audio to PCM internally before playback. This conversion weakens the format’s core advantage and misleads consumers who pay premium prices specifically for native DSD processing.

However, consumers remain largely unaware that this conversion occurs inside their players.

Here’s why this happens and why it matters:

Why SACD Players Convert DSD to PCM

A lot of SACD players don’t stay in the pure DSD domain for long. After reading the disc and decrypting the data, the audio typically enters a PCM-based path at various points depending on design. Some use an all-in-one SoC, while others split transport, DSP, and DAC.

But the end result is similar, as PCM is where most of the processing happens.

This choice is usually about practicality, features, and silicon availability, though, and not an attempt to “ruin” the format.

• Licensing and copy protection
• Chip and market reality
• Processing needs (bass management, delays, room EQ)
• Digital outputs and interfaces
• Universal-player economics

Licensing and copy protection

SACD’s Scarlet Book copy-protection and chipset licensing often steer designs onto approved platforms. Unfortunately, many of those handle DSP in multibit/PCM domains.

That’s why when you build around those platforms, staying DSD-native throughout the whole player becomes the exception, not the rule.

Chip and market reality

DAC and media chips are built for the largest markets, such as CD, streaming, video, automotive, where PCM is the common language. True end-to-end DSD hardware exists, but it’s niche and expensive.

Using widely available PCM-oriented silicon lets brands hit price, feature, and supply goals.

Processing needs (bass management, delays, room EQ)

The features people expect, like bass management, channel trims, speaker distance, room correction, and even some volume controls, are far easier (and often only possible) in PCM.

Many designs convert DSD to multibit/PCM, do the math cleanly, then go to analog. Some will re-modulate to DSD afterward, but once you’ve left the 1-bit stream, you’re no longer “pure DSD” end-to-end.

Digital outputs and interfaces

Consumer S/PDIF (optical/coax) doesn’t carry native DSD as part of the standard. Players that offer these outputs must send PCM at a supported S/PDIF rate (model-dependent) or disable the SACD layer.

DoP (DSD over PCM) and HDMI can move DSD, but they rely on PCM-like framing or HDCP handshakes, and both sides must explicitly support it.

Universal-player economics

Universal players also handle CDs, DVDs, Blu-ray, and streaming apps. Their video/DSP SoCs are designed around PCM workflows because everything else in that stack, like menus, mixing, secondary audio, streaming codecs, expects PCM.

Maintaining a separate, fully DSD-native signal path alongside that adds cost and complexity most products can’t justify.

Only a Handful of Players Keep DSD End-to-End

While conversion is the norm for practical reasons, a few designs keep DSD native in stereo/direct mode.

Before we start, let’s define terms up front. Here, “native DSD” means the stereo signal stays in the DSD domain on its way to the analog outputs in the player’s direct/default mode. There’s no detour through PCM for internal processing.

Turn on bass management, time alignment, room EQ, or multichannel features, and most players will switch to PCM.

With that scope set, these models are known for keeping DSD as DSD in typical 2-channel playback:

• Early Sony reference models (SCD-1, SCD-777ES). Launch-era flagships widely documented as taking DSD straight to analog without a PCM step. They’re the clearest historical examples of Sony’s “pure DSD” approach.
• Sony SCD-XA5400ES. Routes DSD directly to its DAC/HDMI path in stereo/direct mode based on available documentation and user testing.
  Sony doesn’t explicitly promise “no PCM anywhere inside,” so treat this as strong, though not absolute, evidence of native DSD in that use case.
• Esoteric K-series (e.g., K-03XD, K-05XD). Esoteric’s Master Sound Discrete DAC is designed for DSD-native handling rather than relying on an off-the-shelf PCM-centric chip. As with modern delta-sigma designs, internal multi-bit modulation may exist. But that is not the same as converting the program to a PCM stream for general DSP.
• Accuphase (DP-770, DP-720; DP-950 transport + DC-950 DAC). The MDSD architecture is built to process DSD without the usual PCM-based detours in stereo playback. Note the roles: DP-950 is the transport; DC-950 is the DAC.

Other players commonly cited for native DSD in stereo/direct mode:

• Denon DCD-A110
• Technics SL-G700M2
• Luxman D-10X
• Yamaha CD-S1000 / CD-S2100 / CD-S3000

For these, manufacturer literature and owner measurements/support statements indicate DSD-to-analog without PCM conversion in the main stereo path when processing is off.

What the Conversion Actually Does to Sound Quality

When DSD64 is converted to properly implemented high-res PCM, what you hear from 20 Hz to 20 kHz usually doesn’t change. The audible trade-offs are small and come from how the conversion is done, not from the idea of converting itself.

What’s preserved

• In-band resolution and noise: DSD64’s practical, in-band SNR is about 120 dB (20 Hz–20 kHz). That’s roughly on par with ~20-bit PCM at 96 kHz. Converting to 24-bit/96 kHz or 24-bit/176.4 kHz keeps the in-band noise floor well below audibility; good converters measure ≤ −130 dB noise and sub-0.001% THD after conversion. In typical systems, your DAC’s analog stage and speakers/headphones dominate the noise and distortion you actually hear.
• Frequency and level accuracy: With a correct decimation chain, passband response stays essentially flat (≈ ±0.1 dB across 20 Hz–20 kHz) and channel balance is unchanged.
• Timing and phase: Proper linear-phase FIR decimators preserve phase relationships and transient timing in the audio band. Minimum-phase options can be used too; differences are typically below audibility at these rates.
• Headroom for processing: Moving DSD to 24-bit PCM creates working headroom for bass management, room EQ, volume, and mixing without stacking noise. Good pipelines add dither at the final wordlength to avoid truncation artifacts.
• Rate relationships: Choosing 88.2 or 176.4 kHz (integer multiples of 44.1 kHz) avoids awkward resampling from DSD’s 2.8224 MHz (64×44.1 kHz) base, minimizing conversion math and potential artifacts.

What’s lost

• Ultrasonic behavior: DSD pushes a lot of shaped noise above ~20–30 kHz. The DSD→PCM step applies a steep low-pass to keep that out of the audio band. Result: you lose response above ~50 kHz and the signature “rising ultrasonic noise” profile of DSD. That change is measurable, but not audible.
• If filtering is poor: Sloppy decimators can leave ultrasonic remnants (often around −85 dB above ~48 kHz) or allow aliasing if stop-band attenuation is weak. Well-designed filters (e.g., > 100 dB stop-band attenuation, clean transition band) remove this.
• Bad gain staging: If the chain doesn’t keep true-peak headroom, intersample overs can clip during EQ or bass management. Good implementations run a few dB of digital headroom and dither at output to keep things clean.
• Format purity (philosophical): After conversion, you’re no longer carrying a 1-bit stream internally. If the goal is purist, end-to-end DSD, any PCM stage breaks that, even if it’s sonically transparent.

How to Get Native DSD Playback

If you want to keep DSD in its own lane from disc to DAC, you have a few workable paths. Each comes with a catch, so here’s the cleanest way to think about it.

• I²S: the most direct path, with a big caveat
• HDMI DSD: great when the handshake works
• DoP (DSD over PCM): DSD packed for the ride
• USB + software players: the flexible option

I²S: the most direct path, with a big caveat

I²S carries bit-clock, word-clock, and data on separate lines, so it can feed DSD straight into a DAC without converting to PCM. Many modern DACs expose I²S on an HDMI-style jack for convenience.

The catch: there’s no universal pinout. Brand A’s “I²S over HDMI” may not match Brand B’s. Some gear offers switchable pin maps or adapter boards, but you can’t assume plug-and-play.

Use this when: you have a player or DDC (digital-to-digital converter) that can output DSD as native I²S and a DAC that can accept the same I²S mapping.

HDMI DSD: great when the handshake works

Several disc players output DSD over HDMI. If the receiving device (DAC/AVR/processor) supports DSD over HDMI and completes the HDCP handshake, you can stay in DSD all the way to conversion.

The catch: if the handshake fails or the receiver doesn’t support DSD, the player will fall back to PCM, which is often at 88.2 kHz or 176.4 kHz, or refuse to send audio. This is by design for licensing and copy-protection.

Use this when: your downstream device explicitly lists “DSD over HDMI” support. Test with a known-good SACD and confirm the front-panel/status readouts show DSD, not PCM.

DoP (DSD over PCM): DSD packed for the ride

DoP wraps DSD bits inside a 24-bit PCM frame so existing digital links and drivers can carry it. The data remain DSD; the PCM frame is just a container.

Typical requirements are:

• DSD64 → 24-bit/176.4 kHz PCM carrier
• DSD128 → 24-bit/352.8 kHz PCM carrier

The catch: every device in the chain must understand DoP markers. If something in the path treats it like normal PCM, you’ll get noise or a fallback to PCM.

Use this when: your DAC says it supports DoP over USB (common) or over certain digital inputs.

USB + software players: the flexible option

Computer playback gives you the most control. With a compatible USB DAC, you can:

• Bitstream native DSD (ASIO/WASAPI/Core Audio paths, depending on OS and DAC), or
• Choose conversion to high-rate PCM when you want DSP such as room EQ, crossfeed, or volume leveling.

Popular apps (e.g., JRiver, foobar2000 with SACD plugin, Audirvana, Roon, HQPlayer) let you pick per-album or per-zone behavior, set low-pass filters for ultrasonics, and see at a glance whether you’re sending DSD, DoP, or PCM. This is also the easiest way to verify what your DAC is actually receiving.

Use this when: you want native DSD for purity, but also want the option to switch to PCM for processing without changing hardware.