DSD blasphemy

this is blasphemy that may bring on chastising commentary, I realize, but…

I fully embrace the technology of DSD and am intellectually convinced of its superiority. Octave’s The Art of Hifi (Bass) is superb for presenting bass, my primary instrument; non bass is far away, though sometimes this is intentional.

In general, my ears have a different view of DSD. They hear better blackness and purity of some of the sounds, as intended (vocals and piano do quite well). However, well-produced regular streamed content sounds significantly better in most other ways beyond fine/powerful vocals/bass. I have listened to hundreds of songs the past two months.

I have purchased several DSD files from Octave, NativeDSD, and HDTT.

The dynamic range of DSD is staggeringly large, presumably due to less compression, but so much so that to hear the ‘soft’ passages, the volume level becomes too great for the ‘loud’ passages. (Straus’ Also Sprach is my best example) I have to keep my finger at the ready for the remote’s volume buttons.

I was particularly disappointed today by Octave’s Temporary Circumstances. The singer is wonderful and penetratingly clear but overpowering of everything else but most of the robust bass playing. Piano, drums, etc were far away in terms of volume. I dared not turn up the volume to better hear them, the vocal would have blasted me out of the house (likewise for Also Sprach). Under normal circumstances, I would request refunds for my DSD purchases.

In contrast, streamed content like Lovett’s Joshua Judges Ruth (CD quality) and many other performances are wonderful. I find substantial content that sounds good and more satisfying that just about all the DSD I have listened too (my post is already too long to list dozens of great non-DSD performances). My vinyl is always fantastic. Some notable streamed content includes Marcus Miller, Bob Dylan, and many audiophile playlists I found in this forum. All this is great…until I play DSD, which serves in my case as content to study for its benefits (eg, blackness) but wince at its incomplete plate and feeling of being shortchanged.

I generally find that the well-produced stream content plays quite nicely at my amp volume of 25-35, sometimes 40, but that DSD requires volume of 55-65, and even higher for the soft passages (to hear the beginning of Also Sprach, I need a volume of at least 70, but it is still faint, barely audible).

Amp, McIntosh. iFi NeoStream is set to full volume (its DAC performance exceeds that in my McIntosh pre-amp and the newest Bluesound Node. I would never listen to non-DSD content at the required DSD volume levels. For clarity, I am not trying to spare neighbors; they live hundreds of feet away. I don’t particularly mind loud music, though do not need or desire to peg the meter.

Frankly, my system reminds me of what I hear from the best of leading edge high-end equipment since the 1970s. I am gifted with great staging, depth, resolution, presence, and subtlety. Thus, I cannot at this point blame my setup. I presume an Aurender, dCS, or MSB would do better, but I am not convinced the difference would totally explain what I don’t hear in my system with DSD.


I hear ya’ Having found the same issue as well…

One album I was listening with great dynamic range, purposely was playing in the 40is db level
one very quiet section dipped to 20 db then crescendos into the low mid 70 dbs …
(i was using american recorders spl meter).

Had the volume not been turned down…my ears would ave taken a beating…

Not sure of there is a solution for this…

Best wishes


Hi @akro, you clearly have a conundrum here. Yes, you are correct to note that DSD has a tremendous dynamic range. And uncompressed dynamic range is sometimes not the most compatible characteristic for playback in our home audio systems. We are faced with listening as if to a live performance. As you’ve noted in your well written summary, this lack of natural compression in the DSD format can make it very challenging to hear BOTH the very soft and the very loud sounds.

I wonder… do you hear the same issues when listening to 358.2k PCM files (i.e., DXD)? Most DSD256 files we find have been mastered in DXD and then simply output as DSD256, even where the information says “original recording format DSD256”. So, find in your existing music library a DSD256 file that has been generated from a DXD edit master. My suggestion is to try listening to that DXD and compare it to its DSD256 offspring.

Do you hear the same challenges when playing the DXD? I think you should. If you don’t, then I’d suggest trying to borrow and swap out some of your equipment choices. Or, if you have access to listen to a friend’s system, listen on an alternate system.

Sorry not to have a great answer for you. But here’s at least a starting point to figure out what’s going on. For the record, I don’t experience what you describe in either of our primary system or our secondary system here.

Best regards,


Or buy a stereo compressor (a good one) and start “remastering” the dynamic crange on the fly, might be fun if you like fiddling with gadgets and have spare cash to play with :slight_smile:

I’ll just raise my hand and say I have to gain ride myself with some of my Classical recordings. I hate having to do it.

I will say that listening to the very same audio recordings with MBL Extreme speakers, gain riding is not necessary. On the loudest portions it sounds totally natural, and not at overly loud volume. During the quiet passages it is still totally engaging, no need to turn up.

It’s something I really hate about them.

It is one of the reasons I ordered a pair. But I really doubt I’ll go through with it. I need to gain ride on my system maybe one in fifty recordings I listen to. And my current gear is almost completely paid for. I owe a friend for my subwoofers, and another friend for some speaker cables. So easily less than $50k.

(They are really nice speaker cables!)
(and the subs seem nice)

1 Like

Thanks for a great post.

I bought a DSD DAC in early 2016. One download seemed worthwhile, the other three not. So I sold the machine. In most cases DSD files were PCM at some point. The numerous limitations and problems of recording in DSD are well known.

The real problem is you can’t stream it and you have to be prepared to listen to ballet music by long forgotten Norwegian composers rather than the music you’d listen to if you weren’t focused on format.

I have experienced DSD recordings in PCM that are mastered to show off the format’s dynamic range and end up being poor recordings.

It is only blasphemy if DSD is some form of cult, because DSD is an irrelevance to the vast majority of the music listening public. When I first bought PS Audio products about 10 years ago DSD had already died as a format (ie SACD), I tried DSD downloads as I never tried SACD.

My favourite source of purchased music, Linn Records, were pioneers of DSD recordings, they did hundreds of them, they now do exclusively PCM.

The reality is that mainstream artists will not tie themselves to a format that almost no one uses. It would kill their recording careers. The only exception I can think of is Channel Classics. So you get quite a few small labels using DSD for local musicians and a small local audience. There are no such labels here in the UK, where recording in DSD is non-existent. There is one in Germany (Myrios) that has some artists I like and have done some great recordings.

1 Like

Thank you all.

@davida and others experience what I report.

I have decided not to “gain ride” as @aangen aptly notes, not purchase MBL Extreme speakers (though admired but spending hundreds of thousands for them with appropriate equipement is not happening), nor constrict the performance with a compressor (as @joma0711 advised, “start “remastering” the dynamic range on the fly”) to make DSD files bloom in my system the way all other file types do (which is as resolving a system as any I have heard, though with less vertical stage of what I consider great speakers, Magnapan, IRS, MBL, Clarisys…undoubtedly there are others that I have not heard or would pay for, eg Magico, Wilson). Thank you @stevensegal for your insight. Purchased DSD files should not require massive expenditure.

I finally learned not to chase enigmas, almost always the tantalizing promise is unfulfilled.

Notably, my SACD discs do not need the gain riding or compressing of my purchased DSD files. Also, my vinyl does not pose the ‘problem’. My test: Straus’ Also Sprach Zarathustra, Fritz Reiner, Chicago Symphony, either Mobile Fidelity’s pressing of Reiner’s 1954 release or RCA’s Living Stereo 200 gram pressing (both having large dynamic ranged and played with Koetsu Onyx, SME 309, Mitchell Gyro Dec, and Nordost cables to pre-pre-amp). On the purchased DSD, the best parts are virtually inaudible. To my ears and resolving setup, the DSD version represents significant distortion of the performance’s intention.

I will keep listening of course. Maybe Suncoast Audio will open its always-closed ‘store’ to permit auditioning purchased DSD files on what may be the required equipment to embrace them. I value @Rushton’s counsel as well as his excellent reviews of recordings that he shares with us often.

1 Like

One DSD 256 download from Octave at $39 is about double my monthly Qobuz streaming cost. The vast majority of my purchased downloads are from Linn and Hyperion, they are local labels recording world class artists I know and see live on a regular basis, and they cannot be streamed. I pay $15 to $20 per HD PCM file.

It is hard enough for an independent classical label (and DSD is mostly classical) to operate profitably and not sign up to streaming services, without messing with DSD.

I spend far more on vinyl, simply because buying vinyl is fun, and there is lots of great vinyl around. Currently listening to Max Richter on DGG vinyl.

1 Like

I usually use the remote for gain riding in case of very dynamic music independently of format and dynamic range (each format has enough dynamic range to make it necessary for me). I listen quite loud, so it’s usually not because the loud levels get annoying, rather because I often want to hear longer very low level passages a bit louder. Short, full dynamic contrasts are very welcome anyway. I’d do the same at a live event if I could.


I should mention I prefer to keep my peak volume level at around 85db. My old ears can’t take much louder than that. The new amp and preamp allow this as they are sufficiently engaging even at those lower than some of you levels.

Plus I have a bunny who is just getting used to nightly concerts. She is frightened by big changes in volume. Me too.


Ok…when you wrote about the MBL making it possible, I thought of levels around 120dB :wink: 85 dB without compression effects should be possible with most speakers. But yes, the better the gear, the better low level engagement. But if we take live as a reference here, I must say I have more problems with the low levels there than in front of the stereo :wink:

I think with the Extremes we were keeping the peaks around 95dB.

What are the names of the USA labels recording and issuing DSD?

Octave Records, eg new Temporary Circumstances release.

Don’t know where NativeDSD or HDTT are.

As one who records and produces CDs of symphony orchestras I find this thread fascinating. Classical recordings rarely have been exposed to anything other than gentle compression and/or a small bit of limiting.

Also interesting is reading complaints of excessive dynamic range in the context of the loudness wars.

I like the suggestion of purchasing and learning to use a good compressor. Adding overall compression is exactly what the mastering engineer would have done if they thought the dynamic range excessive.


Me too…otherwise it’s just too much for the room (and the “problem” is not necessarily limited to classical music).

No big deal…

1 Like

Also true…

1 Like

begging the question…what is a good compressor?

DBX286s suffice? (this is used for microphones)

The DBX unit will not do the job. You need a unit which handles line level signals and is much more capable and transparent. A good unit will run $4000 or so.


Also true!

I think what should not be mistaken, is, that I guess we all prefer as much dynamic range as possible on the media. We just talk about the need to do some own gain riding in the listening room to adjust the levels to our momentary preference (even when listening to music from vinyl). I think we all don’t want an initial limitation, but do it ourselves if needed.