CD vs HD download question *

Discussion in 'Audio Hardware' started by pez, Dec 20, 2017.

Thread Status:
Not open for further replies.
  1. Ham Sandwich

    Ham Sandwich Senior Member

    Location:
    Sherwood, OR, USA
    Daft Punk's "Random Access Memories" is one example where I hear a difference between the CD and high-res. The sound of "Random Access Memories" has a sort of pillow around the overall sound and also around some of the individual sounds. That pillowy sound is due in part to the style and methods of compression they employed. The high-res version has a larger and fluffier pillow at the edges of the soundstage and around the sounds. It's subtle. But it's there. In order to hear that pillow you first need gear that is able to reproduce and present that pillowy sound well. Not all gear will do it well. Gear that does that pillowy thing well tends to be a bit on the subjective side rather than the pure objective side of sound. If you have gear that does it well you'll hear more of that pillowy sound with the high-res version than the CD version.

    I can hear the difference with my headphone gear. My brother is also able to hear the difference. I asked him to listen to the two versions. I didn't tell him what was different or coach him on what to listen for. And he heard it. The difference is there if your gear is good enough to present it.

    I find "Random Access Memories" to be a useful demo example to have in both CD and high-res. The value factor for the two versions is more difficult to justify just for sound quality differences. The high-res is $18 at HDTracks and the CD can be purchased new from Amazon for under $6. Are you getting $12 more in sound quality with the high-res version? You don't buy high-res for value or if you're trying to control your music budget. High-res is worth it for me for a few albums, but I cannot do a large part of my library in high-res. I just cannot afford it.
     
  2. ralf11

    ralf11 Forum Resident

    Location:
    Earth
    Highland, yes #2 can be an issue -- if you cannot hear a difference then the SQ is not improved

    but if you CAN hear a difference, then SQ may be better or worse


    Overall, there is no generalized solution as to whether to get HiRes (SACDs, HD Tracks or whatever) or not; it depends on how the recording, processing, etc. was done for each album. That's why I came here -- to find out what people think about different releases on CDs, etc.
     
  3. Higlander

    Higlander Well-Known Member

    Location:
    Florida, Central
    And what is good enough?
    And how are you sure you heard same exact mastering's?

    Unless one takes one file and down rezzes it, there are other variables in the equation.
    Not saying you are wrong at all, but that is not a definitive test either, that would hold up as for sure proof.
    If the differences were easily heard, I doubt most audiophiles would even be arguing this.

    I have read (on other audio forums) where guys with quite good gear were not able to discern differences also.
    So I see it as a potential sound improvement that is not proven beyond a reasonable degree still.
     
  4. Ham Sandwich

    Ham Sandwich Senior Member

    Location:
    Sherwood, OR, USA
    The headphone system my brother listened to consisted of a Cavalli Liquid Carbon amp, HD650 headphones with a balanced cable, and a Schiit Modi Multibit DAC. The PonoPlayer is also good enough to let you hear those differences when used with suitable balanced headphones. My bigger headphone gear at home does it better. Much better. But the Liquid Carbon and PonoPlayer are able to do it well enough that you can hear the difference.

    An example of gear that I don't find good enough would be the Schiit Magni 3 in high-gain mode. However, the Magni 3 in low-gain mode can do it. I did a little mini review of the Magni 3 and in that review I mentioned that high-gain killed that pillowy sound with Random Access Memories and that low-gain was able to do that pillowy sound. The Magni 3 in low-gain and a good DAC and headphones will let you hear the benefits of the high-res version of RAM vs the CD version.

    The high-res and CD versions of Random Access Memories are the same mastering for effective purposes. Same EQ, same compression, same levels. In the real world if you want the high-res you buy the high-res. If you want the CD you buy the CD. You don't buy the high-res and then manually downsample it to 16/44.1 to get the CD. In these tests I make sure the CD and high-res are the same masterings (same EQ, compression, levels) but I do not manually resample the high-res to 16/44.1 to test the CD. I use the CD. That's the honest and real-world way to do it. The mastering engineers making the CD have better resampling gear than I do. So I let them do it.

    The difference is heard by real audiophiles. It's the objective gear nerds who argue about it. The audiophiles hear it and just get on with things and buy and listen to high-res without needing more proof.

    You aren't going to get objective proof that people can hear a difference because that sort of objective proof is very very difficult to do. Physiologically and cognitively the brain is not wired to allow people to do full proper ABX testing with those sorts of sound quality differences where the differences are subtle and similar. The brain quickly starts to hear both samples as sounding more and more the same as you got back and forth between the samples repeatedly. You'll have to figure out another way of getting objective proof if that is what you need. I'm happy with the proof I have from my listening.
     
  5. ralf11

    ralf11 Forum Resident

    Location:
    Earth
    actually, no

    numerous scientific studies show the brains is very, very good at making quick comparisons - and not just on audio either

    if you have any studies to the contrary, please give the citations as I'd like to see them
     
    Higlander and Grant like this.
  6. Ham Sandwich

    Ham Sandwich Senior Member

    Location:
    Sherwood, OR, USA
    The brain can do quick listening comparisons when you're listening for artifacts. For example, testing for the audibility of pre-ringing artifacts when developing a lossy codec. That works well. The ears and brain can do those quick comparisons. But for testing the audibility of things that are more about subtle sound quality differences, that sort of quick back and forth listening comparison does not work well and goes against how we hear and process audio. Subtle things like a difference in spatialness, or pillowness, or pinpoint imaging, or slight differences in similar digital filters, and things like that are not things that can be tested and compared reliably using back and forth listening test repeated over and over. With the subtle overall sound quality things, the more you go back and forth and back and forth the more things start to sound alike.

    I'm happy with the proof I have for what I hear with proper high-res. I'm willing to buy gear based on my ability to hear those things I hear with high-res (if the gear allows me to hear what I know is there with high-res then the gear is worth considering, if it doesn't then it's not worth considering). I'm willing to buy high-res music even though it costs more. I'm comfortable making those buying decisions with the proof I have. I'm too old to wait for perfect scientific proof that what I hear is there. I'll be dead or deaf by old age by the time science finds that answer.
     
  7. Higlander

    Higlander Well-Known Member

    Location:
    Florida, Central
    If real audiophiles can hear it, play a random sample to a real audiophile, and they should be easily able to tell if it is high res or not.
    Not sure why that would be difficult.
    What proof are you referring to?
     
  8. sunspot42

    sunspot42 Forum Resident

    Location:
    San Francisco
    24/44.1 is most certainly high-res. In fact, I find the difference between 16-bit and 24-bit audio a lot more dramatic than the difference between 44.1kHz and higher sample rates, particularly anything beyond about 48kHz.

    But mostly it depends on the mastering - if it's a good master it'll sound good on CD and might sound slightly better in 24-bit and at the same or higher sample rates, assuming the transfer or recording was done at 24-bits. But if it's a crap master, no bit depth or sample rate will save it, and often the original CD master from 30 years ago (or some other subsequent remaster from 10 or 20 years ago) blows it away.
     
    daca and Rolltide like this.
  9. sunspot42

    sunspot42 Forum Resident

    Location:
    San Francisco
    How many thousand times does this need to be repeated on this forum?

    HDtracks does not create masters. They are a retailer. They sell whatever the labels give them. Period. Full stop.

    Complaining to HDtracks about "upsampled" content is like complaining to Tower Records (may they rest in peace) about an LP with an off-center spindle hole. They don't make it, they just sell it, and if you don't like it COMPLAIN TO THE LABEL.

    By the way, unless you're willing to do tons of analysis, it's difficult to say that "upsampled" content is merely "upsampled". The label could very well have done some subtle remastering - a bit of compression, some EQ work, etc. - which is why the material was upsampled in the first place (so the work would be easier to do and more transparent). It may or may not be worth the money to pay extra for the resulting "high res" product provided by the label via HDtracks and other retailers, but just because something was sourced from a 16-bit, 44.1kHz master doesn't mean a 24-bit, 88kHz copy is a simple upsampling. They could very well have done work to it in the high res domain.
     
    Grant likes this.
  10. eric777

    eric777 Astral Projectionist

    Well, this is a different question entirely. Hdtracks sales what is referred to as studio quality and is marketed to be superior to the CD. In theory, this would be true if the labels were more honest about what they are sending them. Hdtracks has little control over what they receive. The labels just send them stuff and hdtracks chooses whether or not to sale it. In other words, whether the download is better, worse, or the same as the CD is on a case by case basis and can not be answered definitively.

    As for the price, I don’t agree with it but it is what it is. Their justification for the price is that it’s a higher quality which again is debatable whether or not it is better. In my opinion, the price is nothing more then trying to milk as much money out of people as possible.

    I personally just take it one album at a time. Some 24 bit downloads I prefer over the CD, while others I do not. Unfortunately no one can answer what’s best for you. All I can say is find out which one you like.
     
    Grant likes this.
  11. Grant

    Grant Life is a rock, but the radio rolled me!

    Would you mind telling us how a digital recording recorded with digital gear has no response above 15,000 kHz?
     
  12. Ham Sandwich

    Ham Sandwich Senior Member

    Location:
    Sherwood, OR, USA
    Proof is my own listening. Which is enough proof for me to go about enjoying high-res and gear that properly does high-res. But won't convince people who need objective proof from others to make their decisions.

    Get the Daft Punk in high-res and listen for yourself on various different gear. Eventually you'll find gear that does the high-res envelope and pillowing thing with the Daft Punk and you'll hear what the high-res version is doing better than the CD version. Once you hear the high-res thing going on with that album learn that sound. Learn to recognize it and learn what styles of gear are able to do that. Most of the gear that does that will be the kinds of gear subjectivists tend to like. The gear that objectivists like doesn't tend to do it well or at all. Get familiar with the sound of gear and recordings that does it. Then try doing an ABX style listening test. And you'll come to understand why that type of testing doesn't work with identifying the differences that you've learned to hear.

    That was basically my journey. I used to be an objectivist before I became active on this forum. I've changed my ways after I heard the light. The subjectivsts were right all along.
     
    Last edited: Dec 21, 2017
    Doug Sclar and Grant like this.
  13. Ham Sandwich

    Ham Sandwich Senior Member

    Location:
    Sherwood, OR, USA
    I wish there was a standard term for the middle ground of 24/44.1 and 24/48. What I hear as a high-res style of sound is more about the sampling rate than bit-depth. The bit-depth does play a part. But to my ears and current gear the sampling rate is more important. But for true high res you can't have one without the other. Bit-depth and sampling rate are tied together. 24/44.1 and 24/48 isn't what I consider true high-res. It's high-res in comparison to 16/44.1 redbook. But ultimately not what I want high-res to be.

    I'm using a Modi Multibit DAC right now in my transportable rig right now. It's a 16-bit DAC that does 88.2 and 96 with the Mike Moffat combo burrito oversampling filter. It does 176.4 and 192 as NOS with no oversampling filter. I don't care much for the 176.4 and 192 NOS sound. I really like the 88.2 and 96 sound. I've been enjoying my 88.2 and 96 high res this evening even though it's just a 16 bit DAC. I've also got some 24/44.1 recordings and while they sound good, they're not doing the style of high-res goodness I hear with the 88.2 and 96.
     
  14. sunspot42

    sunspot42 Forum Resident

    Location:
    San Francisco
    I can answer that - the microphones, consoles, processors or other equipment in the signal chain either couldn't record it or deliberately filtered it out. Or the instruments didn't produce anything audible at those frequencies, anyhow.
     
    anorak2 and Higlander like this.
  15. sunspot42

    sunspot42 Forum Resident

    Location:
    San Francisco
    They're totally independent, actually.

    Well, there's your problem - get a 24-bit DAC if you wanna hear how 24-bit can make audio sound better. No 16-bit DAC is gonna reveal the difference.

    I'm not the only person to have noticed the improvement increased bit depth seems to bring with it, or how it's much more noticeable than increasing the sample rate. Roger Nichols, Steely Dan's proficient engineer, also noted that increasing the bit depth of a digital recording seemed to have much more impact on the realism of the sound being recorded than increasing sample rates - not what he expected. Not what I expected either, but there it is.
     
    Doug Sclar and Grant like this.
  16. Grant

    Grant Life is a rock, but the radio rolled me!

    I knew that, but I wanted the person who made the claim to say it, since he sounded so absolute. But, even so, this does not describe every digital recording.

    No, they aren't, and anyone who works with digital audio knows this. Like Sunny says, they are totally independent of each other. They are two completely separate things. One is about the sampling rate, and the other is about bit-depth.

    Now, what may confuse some people is that many DAW software programs processes them together, but quality programs can do each function independently. The sample rate conversion is always done first.
     
    Last edited: Dec 21, 2017
  17. Ham Sandwich

    Ham Sandwich Senior Member

    Location:
    Sherwood, OR, USA
    Independent, but still tied together for getting the full high-res sound. 24/44.1 isn't going to do it on its own. 16/192 isn't going to do it on its own.

    The Modi Multibit is a 16-bit DAC chip, but is able to do 24-bit playback using some sampling theory math. Similar to how chips like the Sabre are able to do 32-bit and how DSD is able to play 24-bit PCM. The Gungnir Multibit has 19 effective bit resolution. The Yggdrasil has 21 effective bit resolution. They all can play 24-bit. I have other DACs that are able to play a more true 24-bit, but they don't sound as good as my Schiit multibits.

    An interesting thing about playing around with the Modi Multibit and Gungnir Multibit and my other DACs is that I have come to find that the pillowy sort of sound I hear with the Daft Punk in high-res is more due to the 88.2 sampling rate than the bit-depth. The bit-depth factors in to other aspects of the high-res sound. But the pillowy and envelope thing seems to be more due to the higher sampling rate.
     
  18. Grant

    Grant Life is a rock, but the radio rolled me!

    24-bit is hi-rez. Even if your sample rate is 44.1, 24-bit gives you a wider dynamic range, and you can hear the difference. So, no, they do not have to be tied together to be hi-rez.
     
    sunspot42 likes this.
  19. Ham Sandwich

    Ham Sandwich Senior Member

    Location:
    Sherwood, OR, USA
    The extra pillowy thing I hear in the Daft Punk high-res is more due to the 88.2 sampling rate than the 24-bit bit depth. I'd rather have that album in 16/88.2 than in 24/44.1.
     
  20. missan

    missan Forum Resident

    Location:
    Stockholm
    Give a real audiophile a random file of Daft Punk, and he will tell what resolution it has before listening. The need to compare two files is just silly.
     
    Higlander likes this.
  21. Grant

    Grant Life is a rock, but the radio rolled me!

    Are you sure about that?
     
  22. Ham Sandwich

    Ham Sandwich Senior Member

    Location:
    Sherwood, OR, USA
    Reasonably sure. I do think the effect is tied to a combination of both the higher bit depth and the higher sampling rate. But the higher sampling rate has more to do with that pillowy sound than the bit depth.

    I'll have to play around with resampling and converting the high-res Daft Punk down to 16/88.2 and to 24/44.1 to explore this again. Difficult to do now while I'm traveling and only have my Modi Multibit with me. Would be better to wait till I'm back home with my bigger gear to try this again. I've done it before and found the sampling rate to be more important for that effect.
     
    Kiko1974 likes this.
  23. Grant

    Grant Life is a rock, but the radio rolled me!

    :sigh:
     
    missan, eric777 and sunspot42 like this.
  24. Ham Sandwich

    Ham Sandwich Senior Member

    Location:
    Sherwood, OR, USA
    I played around with the Daft Punk high-res a couple years ago to explore what part of the high-res contributed to the pillowy sound. It's difficult to remember details from several years ago. But I remember the sampling rate being more important to me for getting that pillowy sound.
     
  25. therockman

    therockman Senior Member In Memoriam


    And that answers the question of price HOW?
     
Thread Status:
Not open for further replies.

Share This Page

molar-endocrine