As more music is recorded in either native DSD or high res PCM, it would seem to me that (a) the fidelity boost from upsampling would decrease, and (b) slightly less obvious, the software complexity to process upsampling will decrease, as it is easier to extrapolate missing data points from a higher resolution starting point, e.g. 192/24
This suggests there may be numerous high quality upsampling options, at very different price points than dCS and MSB in the future. Thoughts?
The question would be how one convert the old analogue masters into âhi-resâ
New recordings are fine
Some people just revert back to reel-to-reel open tape as they perceive this is as close as you can get to the original recording.
For upsampling⌠even for the existing arrangements there are many ways such as via Roon or let dCS stack doing the magic work.
So I donât think upsampling will fade away. Think of the early digital recordings. Just in which domain the upsampling should come in - I.e. at the remastering of the originals or at the end point?
You raise a good point RK; with native highrez sources, the whole standalone Upsampler equation does change quite a bit.
That said, 99% of mainstream music is still released in redbook resolution, or at most in 44.1/24. So, unless one has a very narrow set of source they listen to, I think itâs safe to say a dedicated Upsampler has a long lifespan yet.
Even Chord Electronics, for example, is developing a next-generation M-Scaler upsampler aligned with their flagship Dave DAC, superseding their current M-Scaler which pairs with the lower range DACs.
The amazing thing to me about the Rossini DAC and clock is how fantastic well-recorded redbook files now sound. In my system, the quality of the original recording is far more important than the resolution.
Interesting to me, the M scaler does not appear to support Ethernet input⌠(Am I wrong?)
As an engineer (albeit, a poor one), I always gravitate to simplicity. I believe audio engineers should have a view to what is best, and just force their will through their products, i.e., (if true), âWe have determined that Ethernet delivers the best digital audio experience, therefore going forward we will only have one input source, Ethernet. We arenât supporting any USB or other inputs. We are sorry. Bye!â
This would simplify the architecture and electronics in the products (and the rear panel, considerably), without sacrificing quality, and then just direct the saved dollars to better components elsewhere in the productâŚ
Just MHO
Youâre right, the Hugo M Scaler does not have an Ethernet interface. In fact, none of Chord Electronicâs products have an Ethernet interface (the only exception of sorts is the Poly which has WiFi, but itâs only compatible with the Mojo 1 and 2).
I donât expect âDave M Scalerâ (or whatever theyâll eventually call it) will have an Ethernet either. Designer Rob Watts seem focused on further increasing the number of filter taps, and I guess possibly adding configurable equalisation capability into his DSP engine (like what he did for Mojo 2).
Unlike dCS, I donât believe Chord has any Ethernet/TCP stack expertise for music streaming over the network, theyâve stuck with USB (and jumped through engineering hoops to provide galvanic isolation for that USB interface 
).
One of the reasons I moved from the Dave/M Scaler to Rossini was to get an ethernet interface. So much better than trying to clean up a USB connection. Not to mention the Rossini is a Roon endpoint.
Should be pretty easy for Chord to contract one of the many existing suppliers of streaming boards. For example, StreamUnlimited, Engineered SA, or ConversDigital; these 3 likely own 99% of streaming boards in high-end streamers/DACs.