High-resolution audio means a sample rate of 96 kHz or higher at 24-bit depth, extending beyond the CD standard of 44.1 kHz/16-bit. Apple Music, Amazon Ultra HD, and Tidal all now deliver 24-bit/192 kHz lossless streams via FLAC or ALAC after Tidal completed its removal of MQA in July 2024 (What Hi-Fi?, 2024). Getting that signal from your phone to your speakers at actual hi-res quality depends on every stage of the chain: source, DSP, amp, and speakers. Most car installs lose it well before the amplifier.
- Apple Music Hi-Res Lossless, Tidal Max, and Amazon Ultra HD all deliver 24-bit/192 kHz via FLAC or ALAC. Tidal dropped MQA entirely on July 24, 2024 (What Hi-Fi?)
- In a 2016 AES meta-analysis of 12,000+ blind trials, trained listeners correctly identified hi-res audio at ~60% accuracy, above the 50% chance baseline (JAES, Reiss 2016)
- FLAC at 24-bit/96 kHz runs 50-80 MB per track, roughly 45-65% the size of uncompressed WAV at the same resolution, with zero quality difference on playback (What Hi-Fi?)
- Most factory head units output at 16-bit/44.1 kHz regardless of the file format. A hi-res file played through a standard head unit is down converted before it leaves the source.
- The Goldhorn P2 DSP Pro V2 handles PCM up to 32-bit/768 kHz and DSD256, with 6 TB onboard storage. Audio Intensity is the exclusive US importer of Goldhorn DSP.
What Makes Audio "High-Resolution"?
Two numbers define hi-res audio: bit depth and sample rate. Sample rate is how many audio snapshots the system captures per second. Bit depth controls how precisely each snapshot is recorded, which translates directly to dynamic range. The industry standard definition is a minimum of 96 kHz/24-bit, exceeding the CD standard of 44.1 kHz/16-bit that's been the baseline since 1982.
The Nyquist-Shannon theorem explains why sample rates matter. To capture a frequency accurately, your sampling rate needs to be at least twice that frequency. CD audio at 44.1 kHz captures up to 22.05 kHz, which covers the 20 Hz-20 kHz range humans can hear, with a small guard band for the anti-aliasing filter. So CD technically covers human hearing. (Wikipedia, 44,100 Hz)
What 24-bit actually buys you is headroom. 16-bit gives you 96 dB of dynamic range. 24-bit gives you roughly 144 dB. In a car audio context, that headroom matters during DSP processing. When you apply EQ corrections digitally, each boost operation brings the signal closer to the processing ceiling. On a 24-bit system, you have around 48 dB of working room before that becomes a clipping problem. On 16-bit, you have none.
Audio Format: Sample Rate Comparison
Why FLAC Became the Dominant Hi-Res Format
Three lossless formats handle hi-res audio: FLAC, WAV, and ALAC. They're not interchangeable in terms of file size, but they're identical in terms of audio data. All three decode to the same raw PCM output on playback. The difference is compression and ecosystem.
WAV is uncompressed PCM. It's the same data format a studio records in during tracking. A 4-minute song at 24-bit/96 kHz runs about 150 MB as WAV. FLAC applies lossless compression to that same data, bringing the file down to roughly 50-80 MB (45-65% of the WAV size) with zero information lost. When you play a FLAC file, it decompresses back to the original PCM before the DAC ever sees it. (What Hi-Fi?)
ALAC is Apple's lossless codec. It decodes to the same PCM data as FLAC. File size difference between the two is 1-3%, which is negligible in practice. Use ALAC if your playback chain is entirely Apple hardware; use FLAC everywhere else. For car audio, FLAC is more universally supported across head units, DSPs, and media players.
MP3 is a different category entirely. It uses perceptual coding to permanently discard frequency information the algorithm predicts you won't notice, particularly content above 16 kHz. A 4-minute track at 320 kbps runs about 10 MB. That data isn't recoverable. You can convert MP3 to FLAC, but you're just wrapping lossy data in a lossless container.
| Format | Type | 4-min track at 24/96 | Decodes to |
|---|---|---|---|
| WAV | Lossless, uncompressed | ~150 MB | PCM (original data) |
| FLAC | Lossless, compressed | ~50-80 MB | Identical PCM |
| ALAC | Lossless, compressed | ~52-82 MB | Identical PCM |
| MP3 320kbps | Lossy | ~10 MB | Reconstructed audio, data permanently discarded |
Source: What Hi-Fi?: Audio file formats explained
What Happened to MQA: Where Hi-Res Streaming Stands Now
Hi-res playback has moved decisively to open lossless formats. MQA's proprietary licensing model is now out of the picture.
MQA (Master Quality Authenticated) was a proprietary codec that charged streaming services a per-stream licensing fee and required a licensed DAC chip to perform a full "unfold" decode. The audio engineering community debated its technical claims for years. The business collapsed in 2023.
MQA Limited entered UK administration (the equivalent of Chapter 11 bankruptcy) on April 3, 2023, after losing investor backing. (Billboard, 2023) Lenbrook Group, the Canadian parent company of NAD, PSB, and Bluesound, acquired MQA's IP assets for $125K total in September 2023. (Audioholics, Strata-gee)
Tidal had been MQA's largest streaming partner. They began transitioning to hi-res FLAC in mid-2023 and completed the switch on July 24, 2024. No MQA catalog remains on Tidal. (What Hi-Fi?, 2024)
Where that leaves the major services right now: Apple Music Hi-Res Lossless delivers up to 24-bit/192 kHz ALAC, Amazon Music Ultra HD reaches 24-bit/192 kHz FLAC, and Tidal Max delivers up to 24-bit/192 kHz FLAC. (Apple Support, What Hi-Fi?) All three use open formats with no licensing overhead on the hardware side. If your DSP or source player was purchased specifically for MQA decoding, that feature is now irrelevant. Any device that plays FLAC or ALAC at 24-bit/192 kHz is current.
Can You Actually Hear the Difference? What the AES Research Found
This question has more actual research behind it than most audio debates. Dr. Joshua Reiss at Queen Mary University of London published a meta-analysis in the Journal of the Audio Engineering Society in 2016, pulling together 80 publications, 18 studies that met statistical inclusion criteria, 450 participants, and over 12,000 individual listening trials. (AES, J. Audio Eng. Soc., Vol. 64, No. 6, 2016)
The result: trained listeners correctly identified hi-res audio at statistically significant rates above chance, approximately 60% correct where 50% is random. The effect was real. It wasn't dramatic. Untrained listeners showed weaker results. The audibility advantage was also content-dependent: recordings with wide dynamic range and complex high-frequency content showed stronger effects than heavily compressed material.
The Reiss meta-analysis is the most credible data point in this conversation, but it comes with a condition that most reviews skip: the audibility advantage was only consistent when the full playback chain supported hi-res. Studies using hardware that down converted at some stage showed smaller or no effects. In a car install, that means the research supports hi-res, but only if your DSP's internal processing rate matches the source file's sample rate. If your DSP is processing at 48 kHz internally and your file is 192 kHz, you're hearing the same 48 kHz output regardless of the source.
The practical framing is this: hi-res audio is audible to trained listeners under controlled conditions with appropriate hardware. It's not a night-and-day difference the way moving from a blown speaker to an intact one is. The gains compound with the rest of the system. A well-tuned DSP, proper time alignment, and clean power will do more for your sound than format alone. But if you already have those things sorted, the format matters.
Where a Car Install Loses Hi-Res Quality
The signal path from phone to speaker involves at least four stages, each of which can cap or degrade hi-res quality.
Most installs lose hi-res quality before the signal reaches the DSP. The weak points are specific and predictable.
The head unit's DAC is almost always the first bottleneck. Factory radios and most aftermarket head units cap their analog outputs at 16-bit/44.1 kHz regardless of the source file. A 24-bit/192 kHz FLAC file played through a standard head unit is being down converted internally before the signal leaves the RCA outputs. You're feeding the DSP a 44.1 kHz signal no matter what was on the USB drive. The fix is a hi-res capable head unit with a 96 kHz or 192 kHz output spec, or bypassing the head unit entirely with a source player that connects directly to the DSP.
The DSP's internal processing rate is the second choke point. Some entry-level DSPs accept a 96 kHz digital input but run their processor core at 48 kHz internally. Check the DSP chip spec, not just the input spec on the box. A unit built around the Analog Devices ADAU1452 processes at 48 kHz. A unit using the ADAU1463 can process at 96 kHz or higher. Those aren't marketing differences. They determine what resolution your EQ and crossover math is actually running at.
The amplifier is rarely the bottleneck. Amplifiers are voltage amplifiers. They amplify whatever comes in from the DSP's output, and bandwidth limitations in car audio amps are well above 20 kHz. Get the source and DSP right first.
Goldhorn DSP: Native Hi-Res Playback from Source to Output
We're the exclusive US importer of Goldhorn DSP. I've been running their processors in my own truck and in customer builds for the past several years. The reason I started importing them is that the specs at each price point beat what was available domestically, particularly for competitive SQ builds where the DSP's noise floor and processing resolution matter as much as the tuning.
Three models in the current lineup are directly relevant to hi-res playback.
P2 DSP Pro V2: File-Based Hi-Res Playback
The P2 DSP Pro V2 is a combined DSP processor and media player. It accepts up to four USB drives for a total of 6 TB of onboard storage and plays DSD256 natively (DFF and DSF file formats) alongside PCM up to 32-bit/768 kHz. This means you can load a hard drive full of hi-res FLAC files, connect the P2 directly to your amplifiers, and bypass the head unit entirely. No head unit DAC in the chain. No 44.1 kHz down conversion.
The DSP section runs an Analog Devices ADAU1463 at 32-bit/192 kHz. The ADC uses dual PCM1840 chips rated at 32-bit/192 kHz for the analog inputs. Output is 12 channels at 4 Vrms. Time alignment resolution is 0.02 ms per channel. Crossovers run up to 48 dB/octave with Butterworth, Bessel, and Linkwitz-Riley filter options. Channel separation is rated below -120 dB. Wi-Fi app control and Bluetooth 5.0 with aptX-HD and aptX-LL are both included.
I run the P2 DSP Pro V2 in my own truck. The practical difference you notice first isn't some obvious hi-res quality jump. It's the noise floor. The DSP's output is quiet. When you're tuning at competition SPL levels and adjusting a 6 dB cut at 80 Hz, you're not fighting converter artifacts in the measurement. The 0.02 ms time alignment precision matters too: at 3 kHz crossover, that precision lets you nail the phase alignment between the midbass and tweeter in a way that 0.1 ms resolution units can't match.
DSP10: 10-Channel Standalone with High Output Voltage
The DSP10 is a standalone 10-channel DSP built around the Analog Devices ADAU1453. Its differentiating spec is output voltage: 6 Vrms, compared to the 4 Vrms standard on most processors. On a system with multiple amplifiers, that extra headroom reduces the gain structure math and keeps the amp inputs from running hot. Inputs include optical (TOSLINK, capped at 96 kHz/24-bit by TOSLINK's hardware ceiling), analog, and Bluetooth 5.0 with aptX-HD.
Each channel gets 31-band parametric EQ, 0.02 ms time alignment precision, and 0-359 degree phase rotation. SKYATOM 3D surround processing is built in, a proprietary Goldhorn algorithm for builds where front stage width is a priority. Auto signal-sensing turn-on handles the power trigger without running a separate remote wire.
DSPA 1012 Plus: Integrated DSP and Amplifier at 96 kHz
The DSPA 1012 Plus integrates a 20-channel DSP with a 10-channel amplifier in one chassis: 8 channels at 80W and 2 channels at 150W, totaling 940W. The DSP chip is ADAU1463, same as the P2 DSP Pro V2. Digital inputs accept 96 kHz/24-bit. EQ and crossover architecture are identical to the other units in the line: 31-band parametric EQ per channel, Butterworth/Bessel/Linkwitz crossovers up to 48 dB/octave, phase rotation 0-359 degrees, 8 preset memory slots. Operating temperature is rated from -40°C to +105°C, which actually covers automotive use.
| Model | Type | Hi-Res Spec | Channels | DSP Chip |
|---|---|---|---|---|
| DSPA 406 | DSP + amp | Standard | 6-ch DSP / 4-ch amp | ADAU1701 |
| DSP10 | Standalone DSP | 96 kHz optical in / 6 Vrms out | 10-ch | ADAU1453 |
| DSPA 1012 Plus | DSP + amp | 96 kHz/24-bit digital in | 20-ch DSP / 10-ch amp | ADAU1463 |
| P2 DSP Pro V2 | DSP + media player | 32-bit/768 kHz PCM, DSD256, 6 TB storage | 12-ch out | ADAU1463 |
| DSP16 Ultra | Standalone DSP | Dual ES9038Pro DAC, femto clock | 16-ch | ES9038Pro |
Source: audiointensity.com/collections/goldhorn
The full Goldhorn lineup is at /collections/goldhorn. The integrated DSP/amp units are at /collections/goldhorn-dsp-amplifiers. For system architecture questions or help matching a processor to an existing install, reach us at /pages/contact-us.
For a deeper walkthrough of DSP tuning after you've sorted the hardware, see our DSP settings guide and what is a DSP in car audio.
