• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

384kHz DACs with 192kHz input/outputs

DaNello

New Member
Joined
Feb 26, 2025
Messages
2
Likes
0
Hi, basic question, I hope. I haven't kept up with this for several years.

I have Spotify Connect with up to 320kbps streaming. Streamers and receivers today often have 384kHz DACs, but the optical, coax or RCA connections max out at 192kHz.
Is there any point, at the moment, to buy hardware with DACs over 192kHz? Is 320/384kHz a hoax if you can't get the signal to the endpoint?

Will something like WiiM Ultra receiving 320kbps Spotify audio and delivering to an amp with 384kHz DACs via 192kHz optical be any better than Spotify going to 192kHz DACs and connections?

I'm just getting back into this. At the moment, I just want to hear Metallica's cymbals via Spotify with minimal compressed distortion.

Thank you
 
Theoretically 44.1Khz is all you need
And 192kHz is panty more (and works over most SPDIF/COAX)

But If you want 384kHz you have to go over USB.

But i doubt there is a Quality difference And especially no Audible Quality difference.

Compression distortion if present comes from the audio Codec Compression (if Lossy)
and has little to nothing todo with sample rate.

it is Hypothetically conceivable that there are audio Compression algorithms that perform better in the audible range if the have massive overhead in sampling rate for example 384kHz
But this is highly Hypothetically and not likly.

"Simple" 24bit 96Khz FLAC or WAV is what i would Consider "high end" and will not give you any audible compression. while it is supported by most DACs /PCs
(Unlike 320kbs / 386khz)
 
Last edited:
I have Spotify Connect with up to 320kbps
Normal Audio CD from the 80s is
16Bit * 2 (stereo) *44.1Khz = 1411.2kbps

A good Modern DAC uses 24Bits and maybe outputs 20 effective bits. at lets say 192Khz this would be 9216kbps

in other words 320kbps is not a lot but maybe enough for compressed 44.1khz to give almost CD quality
 
Thank you all for your quick replies, esp the clarification on units.
 
in other words 320kbps is not a lot but maybe enough for compressed 44.1khz to give almost CD quality
It can in fact give you higher dynamic range than CD. You can input and output 24-bit audio into lossy codecs just fine. Apple pretentiously has a nice marketing name for this: Apple Digital Master. It’s quite trivial actually :) I have no idea if Spotify does this as well, probably not. I also highly doubt that there is any audible benefit to this practice.
 
I don't know what lossy compression format Spotify uses but I assume the masters are usually "CD quality" (16-bit /44.1kHz). The sample rate wouldn't normally change MP3 (which I'm pretty-sure Spotify doesn't use) is limited to 48kHz.

Lossy compression doesn't store individual samples so there's no "bid depth". But the dynamic range capability for MP3 (and probably most lossy compression) is actually greater than what you get with 16-bits.
 
Last edited:
Hi, basic question, I hope. I haven't kept up with this for several years.

I have Spotify Connect with up to 320kbps streaming. Streamers and receivers today often have 384kHz DACs, but the optical, coax or RCA connections max out at 192kHz.
Is there any point, at the moment, to buy hardware with DACs over 192kHz? Is 320/384kHz a hoax if you can't get the signal to the endpoint?

Will something like WiiM Ultra receiving 320kbps Spotify audio and delivering to an amp with 384kHz DACs via 192kHz optical be any better than Spotify going to 192kHz DACs and connections?

I'm just getting back into this. At the moment, I just want to hear Metallica's cymbals via Spotify with minimal compressed distortion.

Thank you
For a quick bit of rundown:

"320kbps" refers to the compression rate of a file or stream. Typically uncompressed data for digital audio is 1411kbps (44100 x 16 x 2 / 1024 - 44100 samples per second, 16 bits per sample, two channels for stereo audio, plus a bit of extra metadata = 1,444,864 bits per second or 1411kbps). Compression brings that down to somewhere between 32kbps and 320kbps for lossy audio, typically.

Hz or kHz refers to the sample rate of audio - how many times per second the analog waveform is sampled. CD-quality audio is 44.1kHz. Most tv/movie audio is 48kHz. Recording studios typically use 96kHz not because it sounds better but because studios will be doing a lot of processing of the audio and capturing more data gives you more headroom.

Bit depth - 16bit, 24bit, or 32bit - refers to how large a "slice" of the waveform is being sampled each time the ADC is sampling the waveform - 16 bits gives you 96dB of dynamic range, 24bit gives you about 120dB of dynamic range, and 32bit gives you almost infinite dynamic range.

CD-quality audio is 16bit, studios typically use 24 or 32 bit audio for similar reasons to using higher sample rates. In theory, sampling at higher frequencies and larger bit depths will lead to the DAC reconstructing a more accurate analog waveform. Some people claim to be able to tell the difference between "standard definition" 16/44 digital audio and "high-definition" 24/96 digital audio. However, in practice, 16/44 is accurate enough to perfectly reconstruct the original analog audio and most tests have shown that if the masters are the same, it's almost impossible to differentiate between 16/44 and 24/96 (or other high sample rate) audio.

A 320kbps Spotify stream is 16/44 and in most cases it's very hard to tell a lossy-compressed 320kbps stream from uncompressed CD-quality audio, but if you have experience in audio engineering and know what kinds of artifacts to listen for, it's possible to hear. If you're just listening for enjoyment though and not trying to prove how good your ears are, you almost certainly won't notice a difference.

No matter how you connect your WiiM to your amp, the WiiM is going to be converting the compressed signal to an uncompressed 1411kbps PCM audio signal. It'll then send that signal to your amp over a digital connection (HDMI, optical, composite) or will further converting it to an analog signal and sending it via RCA.
 
fame argues that higher sample rates are in fact worse than 44.1 or 48kHz:
Only if the 196khz Signals Contains high amount of ultrasonic signals
And
If Your System is susceptible to IMD from Ultrasonic signals.

In realty this is not a problem.
 
Only if the 196khz Signals Contains high amount of ultrasonic signals
And
If Your System is susceptible to IMD from Ultrasonic signals.

In realty this is not a problem.
Sure, but it improves nothing (for end playback anyway, there are some reasons to use higher sample rates for production) and increases risk of IMD, so there's no reason to use it and good reason not to.
 
It's possible devices with DSP , like room correction etc use upsampling . I know mine dose . It mentions benefits, who knows , mentions headroom, reducing the risk of clipping etc it up samples to 32 384.

Personally iv always preferred to play a file in its native form , whatever that is , be it 16 44.1 or 24 96 , but I live with this internal upsampling malarkey. Again for me upsampling and down sampling are just things that *could* do harm and I'd rather they didn't happen.
 
Last edited:
It's possible devices with DSP , like room correction etc use upsampling
Basically, every DSP solution uses resampling of incoming digital data. Almost all DSPs use a fixed sample rate and their own, separate, clock domain. So even if your digital input is running at 48 kHz, and the DSP is also running at that rate, it will still resample because the two clock domains are not in sync. Exceptions might be AVRs, which are vastly more complex than most setups, and they may allow different setups for clocking in some cases. Another exception may be software-based solutions, where it is easy to recalculate the biquards or resample the FIR coefficients to a new sample rate.
 
but it improves nothing
Tahts also not true. Higher sample rates (can) give more effective bits trough noise shaping and dithering.
Look at DSD for an extreme example. its only one bit and a view Mhz

but I live with this internal upsampling malarkey. Again for me upsampling and down sampling are just things that *could* do harm and I'd rather they didn't happen.
But if you do it in the PC you Have full control over it unlike it happens in the DAC internaly
In the PC with infinite time and compute power Up sampling can be better then on the fly real time Up sampling in the DAC


Unrealizable​

As the sinc-in-time filter has infinite impulse response in both positive and negative time directions, it is non-causal and has an infinite delay (i.e., its compact support in the frequency domain forces its time response not to have compact support meaning that it is ever-lasting) and infinite order (i.e., the response cannot be expressed as a linear differential equation with a finite sum). However, it is used in conceptual demonstrations or proofs, such as the sampling theorem and the Whittaker–Shannon interpolation formula.

Sinc-in-time filters must be approximated for real-world (non-abstract) applications, typically by windowing and truncating an ideal sinc-in-time filter kernel, but doing so reduces its ideal properties.[2] This applies to other brick-wall filters built using sinc-in-time filters.
 
Last edited:
Back
Top Bottom