HiFi Subjectivist ¦ Audio Forum ¦ NVA User Group

Fretless wrote: ↑Mon Apr 12, 2021 3:37 pm My Edirol USB interface records at a max 96KHz / 24bit.

No need to stick to CD-standard as playback is via network streamers, all of which can handle these bitrates.

What software are you using? I use Audacity which supports up to 384KHz.

I am using Audacity but the Edirol interface is running at 96KHz. Any higher resolution is done by upsampling in Audacity and I have already discovered that not to be beneficial to audio quality. This could be down to the fact that Audacity is running on a very limited-capacity processor in an old netbook. It is quite possible that a bigger PC with a considerably faster CPU might be a whole different tale.

So far my experience with ADCs and NOS DACs tends to reinforce my opinion that digital audio devices work best when fed with bitrates they can natively process. For instance: Teradak Chameleon sounds happiest when fed with 96KHz/24bit which means that no internal upsampling is necessary inside the DAC for the TDA1543 chips (running at that resolution). Metrum Musette prefers 16bit signals for its R2R chips and the Denafrips Ares II sounds better with 24bit.

There is at least one Chord Mscaler upsampling unit in use on HFS and that is a serious upsampling device that does appear to produce good results. Not cheap though.

Fretless wrote: ↑Mon Apr 12, 2021 9:43 pm

There is at least one Chord Mscaler upsampling unit in use on HFS and that is a serious upsampling device that does appear to produce good results. Not cheap though.

The Mscaler upsamples 44kHz to 705kHz and 48kHz and above to 765kHz through a dual BNC connection to the DAC. If the DAC only has a single BNC input the upsampling is halfed. I have what I believe is the cheapest DAC with dual BNC inputs, the Qutest. It has taken it to a completely different level. The biggest benefit is an enhanved sound stage with a greater sense of seperation. On many recordings which I had become acustomed to hearing as a wall of sound, this is no longer the case. I am convinced Cubes bring out the most in what it has to offer and the listening experience is greatly enhanced. Not cheap, but I have had it for two years and the Qutest for over three years and they are my end game source. Everything is relative and the Mscaler is a fraction of the price of the Blu Ray CD transport which introduced the technology and cheaper than music strangelig WAF speakers I have heard.

Now doing a rip of Joni Mitchell 'Dog Eat Dog'. This is another album that has never had a good CD issue and listening to the vinyl for the first time in 30(?) years I can hear a lot of subtle detail that the digital version does not clearly reveal.



Joni had a bad deal on CD: it took ages before her albums came out in the format and the original Warners edition was poor, the late 90's HDCD discs are far better and Rhino have just released a new box set of her early LP's.

She is one of the Greats and a songwriter of incredible insight and power. 'Dog Eat Dog' is her protest album where she complains vividly about all that was wrong with the world in the 1980's. Trump was lucky she was no longer recording during his presidency - she would have skinned him alive.

My sanity (what little is left of it) is being put under severe strain by the growing evidence that I have been TOTALLY WRONG for about half of my existence in this current mortal incarnation.

Digital isn't better than vinyl (sort of).

I am confused and hope that you will be too. It started out as an experiment to see if I could get a decent digital copy of a couple of LP's - and has now switched on my Audiophile-mania circuit as I have found that these rips sound fundamentally different to their CD-based counterparts. This difference is primarily emotional.

Moving from LP to CD in the late 80's, I was convinced that the little silver jobbies were the future and that vinyl would die a quiet death on the scraphead of forgotten technologies like LaserDisc, V2000, Betamax, DCC, DAT, Minidisk etc. etc. If someone had told me that, 30 years later, CD was a dying breed and vinyl on a resurgence - then I would have laughed out loud. Silly me.

So, vinyl rips. Took a while to sort out my recording interface and getting it to run at its max rate of 96KHz/24bit. The right leads were obtained and eventually I worked out how to switch the TT on again and get that sounding passably decent. Having now made new, hi-res copies of about 20 LP's I can begin to look at the process in a new way.

System restrictions place my turntable in a different room from my main, critical-listening (digital-only) setup and that is where the computerised copies really shine. It is hard to criticise modern CD remastering in terms of sharp-edged detail and punchy dynamics - but it remains limited by the 'Red Book' standard of 44.1 KHz sampling rate and 16-bit resolution. My old, original LP's are all from the pre-digital era when recording, mixing, mastering and pressing were all carried out in the analogue domain, in old-fashioned studios where big reels of multitrack tape were running, rather than PC screens and file storage processors.



"Excuse me - but your vinyl rip is still a digital copy and not the sound of the LP." Er, yes, That is true - but it isn't the sound of the CD either (which is, I think, the point I'm trying to make here).

Also, the music (and these LP's represent the music of my 'formative years') resonates intensely in my soul. These are the albums I grew up listening to, the recordings that have laid the basis and shaped my appreciation for a lifetime as a music-lover. These LP's MEAN something.

The superior sampling-rate of the digital copies coupled with the work I have done to get the utmost out of my domestic network-based music installations gives me the convenience I have grown used to - plus the 'character' that vinyl has over CD. I can't properly describe the sensation - but you hear the occasional click, the run-in crackle, the organic-ness of a stylus in a groove. And I have the idea that the music 'lives' more, instruments and voices inhabit defined places in the soundstage, stereo effects are richer, more solid, the performance and intention of musicians comes through with more clarity.

This is a ramble, my apologies, still trying to work out for myself the whys and wherefores of this, it has gripped me and ignited a passion for music that I haven't had for a while / long time. Just ordered a stack of 2nd-hand LP's via Discogs - all albums that I remember from my youth and carry that nostalgic yearning for old friends you want to meet again.

The emotional connection.

This is how i see digital compared to analogue.
Im quite happy to be corrected on this because i have a very rudimentary understanding of how it works.
Lets take 16 bit 44.1khz as an example.
The 16 bit part influences what frequency bandwidth can be sampled, the 44.1 bit part describes the number of samples per second.

an analogue waveform is one continuous sample that is restricted mechanically and electrically by the sampling equipment (recorder) and media bandwidth.

An analogue waveform will be a smooth curve, equipment limits notwithstanding.

The digital samples are a tiny snapshot of what is going on at the time it turns 'on' and samples for that brief moment then turns off again.

It is on and off for the same amount of time.

So, my logic says, that there is half the information recorded compared to analogue, irrespective of the sample rate.

The higher the sample rate, the more times it turns on and off, so the more samples there are, but it is still off for the same proportion of a second, i.e it is on 44100 times and off 44100 times, or if 24/96, it is on 96000 times and off 96000 times. The proportions are the same.
The difference in sample rates is the fact that they generate more snapshots, and the more snapshots, the closer it is to the original smooth analogue waveform because there are more data points that are closer together in terms of time.
The higher the bit rate, the broader the frequency range it can sample.
An analogue waveform is drawn as a continuous wave, digital is a series of dots in the approximate shape of that wave.
Perhaps this is why some people prefer analogue, because digital essentially fools the brain into hearing a continuous sound by playing more clips than the brain can process individually, whereas analogue sounds natural because it reproduces a continuous wave that the ear and brain is built to accept.
Same as a tv. Its a series of still images displayed in a sequence faster than the brain can individually identify, so we percieve it as being a moving image.
Most tvs are now 60hz, so displays 60 pictures per second, over twice as many as are needed to fool the brain.
Analogue signal is always there, digital signal is always half there. Its the bits between the 1s and the 0s that are missing, and this is percieved, not noticed.

Its not 'something' that analogue has, its 'something' that digital does not have

But of course im quite happy to be corrected if im talking utter bollocks

An interesting perspective Fretless; there are technical differences between vinyl and CD that work in the latter's favour and others that favour the former. CD has a greater dynamic range than is possible with vinyl but it can only be realised if the master is also in the digital domain. Analogue magnetic tape has a dynamic range similar to that of vinyl, depending on track width and tape speed, even at 30ips with companding it still falls short of 16 bit digital recording in this respect. The sampling rate being only 2.5 times the upper end of the audio spectrum is one of the failings of CD.

Unfortunately dynamic range isn't the only consideration. When magnetic tape saturates the effect remains musical, when you hit the limits on digital the result is far from musical so we tend to view analogue recordings as more pleasing. Both tape and vinyl also have the tactile element, we are not divorced from the music in the same way as with digital files, we have had a, minor, hand in its happening. It is hard to describe why vinyl is so much more pleasing than CD but it is. A 24 bit 96 KHz recording from vinyl is a very different thing from a 16 bit 44.1 KHz CD recording. Obviously the full dynamic range of the original performance is lost for ever but hopefully the higher bit depth and sampling rate have captured the essence of vinyl.

Contrary to popular belief you can't just roll off the frequency response of a system at 16KHz, even if many people can't actually hear that high. Take away the higher frequencies and even people with seriously impaired hearing can detect its absence (I've tried it with a friend who was over 90 at the time).

It seems that the industry wants us to believe that Class D amplification and digital recording/reproduction are the future. Richard would be horrified at class D, chopping up the signal and then filtering out the (RF level) artefacts isn't what he would call Hi-Fi. Interestingly, the higher the sampling rate of a digital conversion the closer the output wave form becomes to analogue, lets cut out the middle man and stay analogue all the way through the process.

Once the signal leaves the dac it's analogue again, and our amp and speakers treat it just like any other signal.
So my question is then, is there a number of "dots" (I prefer to think of them as "slices") needed before that information being fed to the amp is indistinguishable from analogue?
Or is there a fundamental difference in the nature of the sound which means it will never be as good/be the same?
My guess is that if the slices are thin enough, (dots close enough together,) then we won't know the difference. But that's a guess.
Also. What about records cut from a digital master? Do they sound worse than all analogue? Do they still sound better than a FLAC file or the like?
DQ made some interesting comments on the Genesis re-masters recently. Maybe the answer is that the processing/mastering process is critical to the finished sound, and that makes the most difference?

Vinyl-ant wrote: ↑Wed Apr 21, 2021 9:51 am This is how i see digital compared to analogue.
Im quite happy to be corrected on this because i have a very rudimentary understanding of how it works.
Lets take 16 bit 44.1khz as an example.
The 16 bit part influences what frequency bandwidth can be sampled, the 44.1 bit part describes the number of samples per second.

an analogue waveform is one continuous sample that is restricted mechanically and electrically by the sampling equipment (recorder) and media bandwidth.

An analogue waveform will be a smooth curve, equipment limits notwithstanding.

The bit depth determines the dynamic range, not the frequency band width. In photography the bit depth determines the levels of brightness, in both cases the more bits the wider the range of levels that can be accommodated before everything becomes the same. For example with CD anything above 96dB will come out as 96dB (a very crude analogy).

The sampling rate is what affects the frequency band width, a higher sampling rate means higher frequencies are captured.

This is easier to describe in pictures than in words.

Could a part of the difference also be the fact that you're recording/digitising, the surface noise from your vinyl? No matter how low it is, it's still always a part of the music, whereas a straight digital recording wouldn't have that. I've always felt (and kept it to myself, because it sounds daft in the cold light of day) that surface noise somehow contributes to the "3D-ness" of a recording, which is one reason vinyl sounds "warmer" than digital.

I'll get me coat.