• 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!

CHORD M-Scaler Review (Upsampler)

Rate this product:

  • 1. Poor (headless panther)

    Votes: 380 88.2%
  • 2. Not terrible (postman panther)

    Votes: 14 3.2%
  • 3. Fine (happy panther

    Votes: 7 1.6%
  • 4. Great (golfing panther)

    Votes: 30 7.0%

  • Total voters
    431
Indeed they do, otherwise - if they read the numbers, at least part of them - there is no way they could justify that price... :facepalm:
Presumably the deposit is refundable, but they obviously hadn’t heard it when they plonked down the dough.


At any rate, Rob Watts can apparently hear mosquitoes fart at over ten light years (-300db), which is indeed well beyond our instruments’ noise floor.

Post in thread 'Master Thread: Are Measurements Everything or Nothing?'
https://www.audiosciencereview.com/...ents-everything-or-nothing.62954/post-1760234
 
At any rate, Rob Watts can apparently hear mosquitoes fart at over ten light years (-300db), which is indeed well beyond our instruments’ noise floor.
Somehow I feel very sorry for him, I do not suffer from this form of extremely-sensitive hearing apparatus, so I can hear my music cleaner than he could ever do...
 
This new Chord Scaler also has an ADC. Rob posted two samples. A recording made with a conventional ADC and the same with his ADC. The Headfi crowd went all gooey over the difference of course. :rolleyes:
 
In my opinion, the “-300 dB noise floor” often quoted for Chord DACs should be interpreted in a mathematical / numerical sense rather than as a physical analog performance figure.

At that level, we are clearly beyond anything achievable in the electrical domain. Instead, it reflects the internal precision of the FPGA processing — essentially the numerical noise floor of the digital signal path.

Maintaining such extreme precision is important for DSP operations like filtering, interpolation, and anti-aliasing, where rounding errors and quantization noise can accumulate. So the goal is not to achieve a literal -300 dB analog noise floor, but to ensure that numerical artifacts remain negligible and do not fold back into the audible range after processing.

In other words, it’s about preserving accuracy during computation, not about the measurable noise floor at the analog output.
 
In my opinion, the “-300 dB noise floor” often quoted for Chord DACs should be interpreted in a mathematical / numerical sense rather than as a physical analog performance figure.

At that level, we are clearly beyond anything achievable in the electrical domain. Instead, it reflects the internal precision of the FPGA processing — essentially the numerical noise floor of the digital signal path.

Maintaining such extreme precision is important for DSP operations like filtering, interpolation, and anti-aliasing, where rounding errors and quantization noise can accumulate. So the goal is not to achieve a literal -300 dB analog noise floor, but to ensure that numerical artifacts remain negligible and do not fold back into the audible range after processing.

In other words, it’s about preserving accuracy during computation, not about the measurable noise floor at the analog output.
In that case, it would still be useless: The advantage would only materialize if the signal stayed in the digital domain and would be processes further. But that's not the case, it is a DAC after all. If there is zero measurable advantage at the analog outputs, then there is zero advantage to this type of processing. It's also not what Cord claims themselves - they suggest it one could even audibly detect such tiny signals/errors, which is beyond absurd.

Just to reiterate: The number Cord claims would be about nine orders of magnitude below the noise floor of the best modern DACs. It is nothing more than bullshit marketing for customers who do not understand the tech.
 
What I can't understand is how anyone could be excited for this new thing after the M-Scaler measurements and review. It's pathological.
 
Maintaining such extreme precision is important for DSP operations like filtering, interpolation, and anti-aliasing, where rounding errors and quantization noise can accumulate. So the goal is not to achieve a literal -300 dB analog noise floor, but to ensure that numerical artifacts remain negligible and do not fold back into the audible range after processing.

24 bit / -144dB is perfectly adequate for this.
 
Just do some controlled blind listening tests and that would be end of debate.
 
Just do some controlled blind listening tests and that would be end of debate.
You don't even need blind tests if you can test them at the same time and are actually honest with yourself.
 
You don't even need blind tests if you can test them at the same time and are actually honest with yourself.
Being honest with yourself doesn't help. We can't turn off our perceptive biases.
 
Being honest with yourself doesn't help. We can't turn off our perceptive biases.
I was honest with myself that I couldn't hear a difference between two different AMP/DACs. There are definitely biases we have on everything. I had a preference for one and couldn't articulate why, even though I could not tell a difference between them. I am sure it's because of the R2R brain washing in the audio world, or the physical form factor.
 
  • Like
Reactions: MAB
Being honest with yourself doesn't help. We can't turn off our perceptive biases.
If I had a dime for every time I heard someone say “I was expecting the opposite” or “I wasn’t expecting anything” and averring that constitutes a proper control.
 
Back
Top Bottom