With video, I’m amazed at how picture processing of TV programmes can dramatically improve the perceived picture quality. I’ve an LG OLED tv (hence UHD) which is renowned for having excellent upscaling firmware - it really works to the point that even SD broadcasts are greatly improved.
i wonder whether it’s possible to process audio in some way - I have a Chord M-Scaler which claims to upscale effectively. I think it does work, to a point - e.g. in a quick test I couldn’t discern a difference between a 44.1/16 v a 44.1/24 being upscaled to 96/24 (I am however suspicious that the 24 bit version may not be genuinly 24 bit - the music was Billie Eilish).
Beef, the Chord M-Scaler is unique in terms of the algorithm that it uses to upscale. It’s the brainchild of a chap called Rob Watts. I bought it to upscale my CD data to 88.2/24 and that goes to a W4S Remedy reclocker before entering the DEQX HDP-5 digital crossover. The crossover outputs 96/24 to the three DSs (I also send the DEQ output through W4S remedy reclockers). The W4S reclockers have femto clocks in them.
Here’s a link to a review which describes the maths behind Rob Watt’s algorithm:
Good stuff @jazznut . Occurs to me that this all gets into some pretty academic sorts of views of the questions - something I wasn’t thinking about at first. In reality, it’ll be pretty rare to use a set of digital conversions in between the performance and the playback that attempts to do NOTHING else to the signals aside from capturing it as “faithfully” as possible. That scenario being where it is easy to see the potential superiority of digital over analog.
However - just as in a fully analog-domain recording and repro chain, you’ll use EQ and so forth in the digital stages to give either the most realistic-sounding result or the most euphonious - or some combination of both.
Certainly analogue processes have their well-known losses and gains, while digital’s losses grow ever smaller over time, seemingly. The thing I find fun and ironic currently is the popularity of digital recording plugins that faithfully emulate classic pieces of analog gear.
This is where any sort of argument about the comparative “perfection”, so to speak, of digital processes inserted into the performance-to-playback chain start to become at best academic, and incresingly less meaningful, IMO.
An excellent read, thanks
It also gives the lie to the oft-repeated criticism of digital that it “only takes snapshots” and “can’t faithfully reproduce the shape of an analogue waveform”. This all assumes our understanding of hearing (and specifically the bandwidth of our hearing cutting off completely at ≤ 20 kHz) is correct of course, but I’m aware of no credible evidence to the contrary.
I’m not sure if I understand you here or you might have misunderstood that with “digital processing inbetween” I just mean the normal production of digital recordings and media between the two analog stages you mentioned (the music recorded and the speakers). No need to go deeper.
Yeah, for me the fun part is that most seem to understand analog “effects, advantages” or “wanted characteristics” as something like “extra rich, extra colored, euphonic, sometimes muted and less aggressive” sound which certainly is producible by plugins.
This for me in contrary is the cliche that would equal defining digital sound as harsh, thin, rather too rich of detail etc., which also is a cliche of avoidable characteristics with appropriate choices of gear.
When I talk of things getting lost in digital versions of the best all analog productions, it’s a certain palpability of sound and space in the performance, ease of rhythm and pace, openness at both frequency extremes, energy/dynamics and holographic imaging. Nothing about any of the characteristics most would try to “reproduce” (or could achieve) with plugins. For me those for the most part are the characteristics to avoid and to enjoy lacking in good digital. You might just need them for bad digital to compensate.
Not sure how fair the comparison is, but at least both of the following releases were made by the same mastering engineer under the same label. The left one issued 1987 as all analog produced release on 33RPM, the second only few years ago produced from hires files on 45 RPM. The first one sounds so obviously better in terms of realistic piano sound in the room, that I tend to think it’s an unfair comparison because there must have been other mastering choices involved.
Sometimes I get this crazy notion that something resembling a conversation can be carried on here, but it is so non-realtime that it doesn’t work often I wasn’t so much commenting on anything you said as continuing on about the subject. Apologies if it seems you were misunderstood.
About the time I phones and Ipods came on the scene…I wish I could recall the forum where
this very interesting post came up…
The jist of it was kid goes to college.,all his college buds thought music played through
these with earbuds was the thing…until the young college boy brought his buds to his house
for the holidays…when he took them down in the basement where his dad’s “old dinosaur rig”
resided…the college kids had no clue as to what this monstrocity was…
Until the dad put on some vinyl Beatles followed by Beach Boys …old stuff …you know…
Turns out they spent their entire holiday in that basement cranking the tunes…
What a story these guys must have had to tell their other college friends…
Needles to say those college kids got a lesson they wouldn’t soon forget
It has been a long time since I read that post…I’ll try to find it and link it
This is a similar function to HQPlayer, which costs a tiny fraction of the M-scaler, not county the device to run HQP. There are posts on the Roon HQP forums of several Chord M-Scaler owners selling it in favor of HQPlayer.
For me, and I own licenses for both, Audirvana and HQP both sound great. For better or worse, HQP has far more filtering and upsampling options. The dealbreaker with Audirvana was the buggy and antiquated GUI/ iOS Remote compared to the green fields of Roon.
I used to use a computer to rip/store music and replay via USB to my crossover. I decided to replace the front-end with the best components (IMO) to get the best possible feed into the crossover. That resulted in me geiing the Melco server, the M-Scaler and a couple of reclockers to feed 96/24 to the crossover (the M-Scaler uses a unique algorithm to upscale).
Sorry about the tardy reply, but things have been hectic around here, and I wanted to find some time to put the Kestrel 2s back in the system for some listening after 9 months now with just the Maggies.
The biggest difference is that the Kestrels have a more pronounced top end. I don’t think it’s more extended, just emphasized a little. A minor goosing perhaps. The second difference I noticed is a very slight compression of soundstage depth with the Kestrels, though I emphasize “very slight.” Same thing - very slight - bloom in the upper bass/lower midrange.
I was surprised to find that tonal balance through most of the midrange and the general overall instrumental timbre are very similar between the speakers. Instruments sound very natural and real. Very clear without being edgy, and with more than decent precision in imaging and sense of air. Transients were quick and punchy.
Given all of that, and the little more forward nature of the highs and lower midrange, the Kestrels are a very seductive speaker, and it probably helps explain why I was so taken with them when I bought them back in the early 2000s, coming from, at that time, Cambridge Soundworks Ensembles (I’d sold my old Maggie MG-1s several years earlier). Frankly, they fared much, much better than I expected after having lived with the Maggies for 9 months. But in the end, I think the .7s are a much more natural sounding speaker. No chance of the kind of fatigue that can set in when some frequencies are prominent, even a little.
I wasn’t so much commenting on anything you said as continuing on about the subject. Apologies if it seems you were misunderstood.
