I’m playing to a Holo May KTE DAC. I sold my DSD senior and DSMP transport a while back. I couldn’t be happier at present.
1 Like
I understand you very well, that using a DSP might not give the desired benefit, “if that is the only way to optimize the sound”. I work with a friend who is also a professional in sound optimisation. We tested over the years and we use a DSP during the very last step in sound optimisation. The DSP part only “nudges” the remaining “sacks” and “bumps” towards the target curve.
As most of us know, sound optimisation starts first with a good set of speakers which are matched with the proper amplifiers, pre-amps and cable configuration setup.
The second step is speaker placement, listening and simultaneous, measurements, and room acoustic treatment and re-measurement and listening. The last one is not an easy one, and also requires investments as well. The process of placement, listening, treatment and measurements is painstakingly long, It is similar to a “growth” process of psychotherapy 
.
To my regret I see a trend that people do not invest in acoustic treatment. main reason is the cost. Most people have the tendency to invest an additional amounts, (say ± 10K) in active components,- next to all the expensive stuff they already have, - then the same 10K in passive acoustic treatment. The paradox to this trend is that the “incremental” gains in clarity and musicality,- once you have systems as PSAudio Stellar Series / DirectStream ,…, - is mainly dependent on acoustic treatment, and NOT buying an even more expensive DAC or pre-amp. (yess, it goes against some resellers interest who claim otherwise)
I’ve run several hunderds measurements in the last 2,5 years, and kept track of the developments, each time a bass or sub trap was added, removed, diffusion was added, absorption, …, you name it. meanwhile playing with speaker placement, Bi-Amp, Dual Mono’s , “Tri-amp” Hybrid as is now, … , and the journey still continues.
So if you would ask me, YES. DSP is a good thing, but only as a “nudge-tool” and NOT as the main tool to optimise sound. If you use it as the only tool, results tend to go to more “closed” “cold” and “dry” soundstage, instead of lushness and clarity and musicality we are all looking for 
.
And when the subtle nudge is right, - software as REW can help you with that , - you will surely hear the audible benefits of DSP, as de mud is gone, and tight low bass is already present by the acoustic treatment. Perfection lies in the way to it, not in reaching it, and like all integrations people make in their life, they take time time one needs to grow.
If you need to give more headroom ,- when configuring DSP,- then -4dB on nominal, then you might consider acoustic improvements in your room or speaker replacement (mine is now reset to -2dB, instead of the -4 a few days ago)
3 Likes
Glad you love the kit…there are a lot of Holo May fans here.
I have to wonder about doing a blanket up-scaling to DSD256 only to have the DS DAC do its thing to the signal anyway.***** The question for me is: Is an up-scaled resolution of a file that is not native to the original file being played/streamed benefit from upscaling prior to being put through its paces in the DS DAC?
*****“DirectStream converts every input signal, both PCM and DSD, to single-bit, high sample rate 20X DSD signal.” [https://www.psaudio.com/products/directstream-dac/]
***** “That means the PS Audio DSD is upsampling the incoming PCM to 20 times the nominal DSD rate (20 x 64 x 44.1kHz) which is 56.488MHz (as in million). After processing, the signal is down-converted to double DSD, sent through a passive low pass filter and out as an analog signal.” [PS Audio PerfectWave DirectStream DAC Review - HomeTheaterHifi.com]
*****“The higher sample rate does not create new information from the original signal but allows for better filtering algorithms which allow for more of the detail of the original signal to come through. Noise is pushed into the higher frequency band and filtered out via a passive low pass filter. This brings out more details and improves imaging.” [ibid.]
Not looking for you to answer directly or defend your practice…rather, I am really wondering about whether taking such steps could render yet more information from a given recording in a beneficial way.
[@tedsmith is a pretty busy guy but maybe I can tempt him to chime in with this indirect invitation?]
Regards.
1 Like
This sounds like one for Ted. I really don’t have a clue. I bought the Holo May because I was ready for a change and it was getting wonderful reviews. The DSD which I got used for $3K was getting old and losing value. It’s still a great product and I might buy another one for my second system.
1 Like
There’s no mathematically perfect upsampling filter. There are multiple upsampling filters with varying compromises. Each user should pick the path that sounds best to him. Sometimes that’s clearly the path that changes the signal the least, but it’s also valid (and indeed likely) that for others a path that changes things a little more sounds better.
A few corrections: the first releases of the DS software upsampled to 10x and back to double rate. later releases upsample to 20x and back to quadrate. For input sample rates in the 44.1, 88.2… double rate the 10x and 20x is immaterial, they get down converted by simply ignoring the other samples so you might as well think of the upsampling of non 48, 96 or 192 signals as going directly to quad rate. 48, 96 and 192k sample rates require more steps to get to quad rate.
The DS has it’s set of upsampling filters and HQPlayer has a broad set of upsampling filters, it’s quite probable that some will like one or another filter from HQPlayer more than the DS’s upsampling filters. I’d even agree that some of the possible choices in HQ player are better (at least in some ways) than the filters in the DS. As far as I’m concerned whatever sounds the best is a good choice for any given person on their system.
As a general rule I’d leave everything in its originally recorded format and then use HQPlayer or the DS (or whatever) to go to quadrate and then to the DS. 2nd best IMO would be to use your favorite tool to go to double rate DSD then to the DS and if DSD isn’t a possibility of your tool I’d go to 176.4 PCM then to the DS. Tho 352.8k seems like a better choice, I like the 176.4 upsampling to quadrate in the DS better. In the DS Mk II if I couldn’t get to DSD with my favorite external tool I’d go to 705.6k PCM.
5 Likes
Thanks, Ted. Seems like “sound” advice…
Scott
Thanks for the insight. I use Roon without HQP and have found I usually prefer any PCM rate upsampled to 352.8 with Roon’s slow rolloff minimum phase filter. Then let the DS do the rest, in to M700s and LS50 metas.
1 Like