Secondary benefits

July 6, 2022
 by Paul McGowan

There’s little direct benefit to me when I go out of my way to help someone. But I feel better when I do.

My good feelings are a secondary benefit.

In the same manner, we can find secondary benefits in such audio-related activities as upsampling.

Upsampling provides no more information than what already was there, yet it can often sound better. That’s because the secondary benefit of upsampling is engaging with a different and better-sounding filter inside the DAC.

Sometimes we’re taking primary actions to enjoy secondary benefits.

It helps to understand why.

Subscribe to Paul's Posts

15 comments on “Secondary benefits”

  1. Unfortunately I don’t understand. Yes, I get the first example, I imagine it’s one of the primary reasons we help others, but not the second. Without getting too technical could you expand a little please. Would it not be possible to use the different better sounding filter without upsampling? A purist might suggest that all manipulations are detrimental in some way.

    1. Rich,
      By way of explanation & as an example…
      …original Redbook (44.1/16) recordings require a very ‘steep’ filter/filtering process when
      turning the digital signal back into analogue.
      This steep filtering messes with the resulting analogue sound, for simplicity’s sake let’s just
      say that it ends-up sounding ‘less musical’ than a digital signal that has been sampled, or upsampled, at a higher rate because being recorded & stored at a higher rate means that
      a less steep (brutal) filter can be employed thus giving a ‘more musical’ sound to the
      resulting music.
      This is a very simplistic explanation (as I’ve indicated) & I’m sure that if I didn’t get it all
      right, either Paul or someone else will chime-in with further details on this matter.

      In short, No, the filtering depends on the bits & frequency of recording digitally.

      1. Thanks Martin, now I understand, a bit more. I was aware of these ‘brick wall’ filters but didn’t get that was what Paul was referring to, if indeed that’s what he was referring to. Imagining instead that PSA had invented some sort of new, magic filter.

      1. Thanks Joseph. I must confess if I was feeling contrary my takeaway from the video would be that “engaging with a different and better-sounding filter” is the primary benefit of upsampling. Paul explains in the post that’s the reason it sounds better. So I understand more now but for me a somewhat confusing post today.

  2. Improved external up-sampling (called D/D converters) goes back a long time and started in pro audio, I think with the dCS 972 in 1997. The dCS Purcell 24/192 upsampler came out in 1999, intended to be partnered with their Elgar DAC. My own Devialet does propietary upsampling to 384/40. PSA obviously also does its proprietary upsampling. The FGPA processors required apparently cost quite a bit of money – hundreds of dollars, compared to say $50 for even the best (e.g. ESS9038 Pro) DAC chips.

    There does not seem to be much love for these external D/D converters. Chord brought out the M-Scaler a few years ago and I remember there being quite a doubt as to their need/efficacy. It was intended mainly to be used with the $$$$ Dave DAC, so it probably had a market that would use it without worrying about the high price.

    There are a few reviews and measurements out there, Goldensound and ASR (today, strangely). They both demonstrate an exceptionally good filter. The processing power obviously generates a lot of jitter, which is not a problem for Chord DACs that are extremely good at noise rejection. It will obviously be a lot more jitter than a basic $100 DAC with much lower processing power, which ASR overlook. It is a false comparison to compare a D/D converter with an D/A converter anyway, but you have to be aware that these devices are inherently noisy.

    Like dCS and Chord, these are external boxes that are essentially part of a system. It reminds me of the Cambridge Audio CD1 from 1985, the first CD player to be split in two into a transport and external DAC, soon followed by others like Marantz, with digital outputs from the standard CD player and matching external DAC. Ultimately most people, including many audiophiles, continued to use standard CD players with the internal DAC.

    The sonic improvements, allegedly transient speed and separation of the image, is subjective. I heard it between the Devialet Le250 and 250 Pro models, with improved filters, but there may have been other things going on.

    Whilst the impact of filters are easy to measure, whether they make an audible difference is an argument that will rumble on indefinitely.

    1. Steven, It is an FPGA ( not FGPA ) which is a Field Programmable Gate Array. Xilinx is the largest maker of these chips ( Xilinx is about to be acquired by AMD ) and their Spartan6 chip is widely used in all kinds applications ( auto, audio, video, etc. ) It is very inexpensive ( decade old 45 nm silicon technology ). A basic module in a quad flat pack will set you back $20, a higher performance one in a PBGA ( plastic ball grid array ) is about $170.

      What is expensive is developing the program for the FPGA. That is where all the innovation is. Right now, unfortunately FPGA’s are being hit by the same supply chain problems that almost all chips are having. This is hitting the automotive industry very hard.

      1. I struggle with acronyms longer than 3 letters, except for CAIOKSSFMOP, which I made up 44 years ago to remember the Citric Acid Cycle. Of course Richard of York gained battle in vain is always handy.

        I am aware of the car-chip issue as Nissan keep on emailing me to ask if they can buy my old banger. They are scraping the bottom of the barrel if they need my car to meet demand.

        What is admirable is that the Chord Mojo2 is the same price as the Mojo I bought 5 or 6 years ago, the price of FPGA chips notwithstanding.

  3. As a side note it is interesting to observe, that e.g. regarding record players or even amps, we come from a generally flawed starting point. Measures to improve the sound are taken against very few basic well known and accepted culprits, more or less only “resonances” when we look for example at record players.

    In digital technology we come from a thinking that we have a more or less flawless concept as a starting point, which up to now is a basis in argumentation, it just didn’t sound like that for a long time. Many measures like the one mentioned today, improved its sound over the years and still do.

    My impression is, it would help to understand what makes this theoretically accurate concept flawed, when we understand what except noise/jitter are the culprits or weak points we work against in digital while improving it since decades.

    The picture, that with digital, we have an accurate and lossless concept since its invention while we communicate more or less trailblazing improvements in regular intervals, seems to lack a certain logic.

    1. The thing is, upsampling can be done with software, but it needs a pretty powerful desktop computer. So higher native sample rates come at a cost of more electrical noise. Plus you need a computer in your system, which I don’t have, never have and never will.

      My experience is electrical noise is the biggest problem for digital audio, but then my unit does A/D and D/A conversion on analogue inputs as well. At least with analogue sources the hardware is primarily concerned with internal noise, that it can control, rather than external noise, which it cannot control, but can address.

      It gets so complex and convoluted, that ultimately, as has has largely prevailed, it will be up to manufacturers to decide what theory works in practise and for consumers to use their ears and wallets to decide what matters.

  4. People in the Southern part of the US must innately know this Paul. Having lived my first 60 years in California, it is clear to me now that we’ve left that wasteland of a society that Southerners stand ready and willing to take advantage of that fine feeling of helping someone else.

    1. Champster, Having lived the first 35 years in the midwest and now ( for about 35 years ) 30 miles north of NYC I get what you mean. I do worry that I have become so disgusted with NYC in recent years where I see only its dark side ( drugs, gangs, violence, danger ) when I use to see mostly its glamorous side ( incredible art, music, food, architecture, etc. ). I worry that the pandemic has turned me dark.

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram