I do think everyone needs to take a step back and respect what David is saying. He hears a difference and he is not alone many other do so please respect this point of view, whether you follow it or not. . There are also lots of you saying David’s wrong with out even listening to the files or trying to understand your just making judgements.
I was interested in understanding why they sounded different to someone when logic and my experience dictated otherwise. I’ve listened and listened and listened and then analysed then listened and listened and came to my conclusion. Please do the same before you decide to make a contribution to this thread. It’s a discussion not a Witch hunt and David himself was keen to work out why they sound better.
Two identifical files cannot sound different. If they sound different, they are not identical. There is something going on that makes them not the same. If they cannot be reliably distinguished in a blind test, they do not sound different.
James,
My first ripping device was the Naim Unitiserve, which I bought in 2013. I guess it cost about €1,300. It did a perfect job, for me. Of course there was a chorus at the time singing that the same result could be achieved via a cheap pc and an external drive.
I never regretted approaching it this way, and I sat for night after night with piles of CDs , putting one in, waiting for it to be ejected and putting in the next. What I am doing now is ripping the same 2000+ CDs via a Melco D100 (£1100?) to a MelcoN10 music store.
There are several ways of skinning this cat, but my experience is that by and large you get what you pay for
If the files sound different it must mean that the electronics is not fully capable to unfolding/unzip the compressed flac without generating noise. Different compression levels between the flac files must be the answer.
There’s no excuse for electronics that can’t effortlessly decompress FLAC in real time. You can’t seriously be suggesting that the CPU inside Rob’s (doubtless expensive) streamer is so lame that it has trouble decoding FLAC?
I could be mistaken but my impression was that Rob thought the more compressed file sounded better.
I listened to these using Roon to my Dragonfly Cobalt and Sony headphones. I can hear no difference. Of course, since I can’t hear them both at the same time, I don’t know how I could tell.
I hope that an honest and generous spirit prevails … lest this discussion go the way of so many others.
What I’d like to understand is this: if David’s two original files sounded so different from each other (at least, to him), how did the 4 that I posted sound? He seemed pretty confident of his identifications and so — if the files he was identifying them with sound so different from each other — doesn’t that mean that there were similarly stark differences among the 4 files that I posted?
Looking at MTBF, feel of the ejection mechanism / tray solidity and its ability to hold the faceplate, ease of changing said faceplate might explain why they’d do something like that.
Yes, this is somewhat plausible. However do keep in mind that the computer works harder (takes a bit longer) to ENCODE higher compression FLAC levels (e.g., level 8), but the DECODING of FLAC files is essentially the same computer work whether decoding a FLAC file compressed at 8 or 1. Many people miss this fact when talking about FLAC decoding.
I appreciate your willingness to participate in these tests as do others. This shows you to be a thoughtful person rather than a “subjective vs objective warrior” as often found in other audio forums, where these discussions never try any tests, turn into name calling, and eventually leave half the participants banned from the forum.
Note that my posts have been mostly about mechanisms whereby you could be hearing differences that are not picked up by the tests (i…e., existence of volume differences caused by RG tags or Roon volume leveling. That is, I’m also trying to have an open mind about how such differences could exist on your end in certain situations.
One mechanism that a couple of people have PM’d me about is the (mysterious to me) discrepancy in the silence-padding of @David_O_Higgins’s files.
For comparison purposes, I removed all the leading/trailing silence in the files. But the silenceremove option to ffmpeg offers a lot more flexibility than that. If you think this might play a role, I invite you to play around with varying the amount of silence retained at the begining/end of the file, to see if that makes an audible difference.
Personally, I am dubious that a 12ms difference in the delay between when you press ‘play’ and when the music starts could make a detectable difference. But, hey, knock yourself out …
The files I sent you differed in 2 respects from the files you sent me
The metadata was removed
The leading and trailing silence was removed.
Otherwise, they were unchanged from what you sent me.
With @Rob_BrownGHF’s files, I didn’t have to perform the second step: removal of leading/trailing silence. He, too claimed to hear a difference between those files. You might try listening to those.
What I was suggesting was that if the leading/trailing silence had some effect on your perception of the files, you might be able to explore that by varying the amount of leading/trailing silence. ffmpeg allows you to do that.
Jacques, I have no interest in fiddling with the files. I took it on faith that what you sent back was the same as what I sent you, but it appears it wasn’t?
I’m not a tech expert, but I really love my music.
David
If they were literally the same as what you sent me then you could instantly have determined which was which from the different track titles. What point would that have served?
.