Sorry but I can help it if you haven't experienced any of the high def formats on a system that can play them adequately. And no I am not talking about a $100,000 system. My system consists of a Yamaha RX-A2A with a Panasonic 4K Player that plays SACD and DVD-A and a half descent set of Klipsch speakers. It is definitely under $1500 and I can very clearly hear the differences between the original CDs and the SACD, DVD-A and Blu-Ray Audio discs.
But no. you aren't going to hear the difference on even the best "computer" speakers nor in most car stereo systems. If your system is only capable of producing 16Bit / 44.1kHz then that is all you are going to hear no matter the source.
Which most people refer to as the resolution. That's why the formats are called "High Resolution".
That's only because your dac isn't good enough at playing CDs.
BTW, DSD64 isn't 64 times better than CD. One CD sample has around 65,000 possible values. The more or less equivalent of that in DSD64 is that there are only 64 values in the same time period. Now of course that isn't quite the right way of thinking about it because PCM encodes a sampled voltage directly. DSD uses Pulse Density Modulation which is a very different thing. Lots of dacs sound better playing DSD content than if playing PCM content. Its not because there is more information in DSD64, its because the PDM tends to sound better than PCM even if it doesn't measure as better.
On the subject of resolution, time resolution all there at a sample rate of 44kHz (up to 20kHz anyway). As I said before there is only one correct solution to recovering the music waveform from the samples. A good dac can resolve everything on a CD, but most dacs aren't that good. Its not that the information for resolution isn't there. Again, the problem is in the dac.
Last edited:
High resolution can also refer to the resolution in momentary value:
CD: 65536 different quantization levels
DSD: 2 different quantization levels
It's all grossly oversimplified, though, because the time resolution of a digital playback chain is not limited to 1/fs and value differences smaller than 1 LSB can be resolved with dither.
CD is exactly 44,100 Samples, (44.1kHz). DSD64 is exactly 2,822,400 Samples (2.8224Mhz) 44,100 X 64 = 2,822,400 64 Times CD.
Given your errors in calculating the sample rates between CD and SACD it follows that this of course will no longer be correct either.
This is central to the thread. Most delta sigma dacs are much better at playing hires files.
You really should read up on the sampling theorem, information theory and logical reasoning.
Just when I thought I was most grumpy member here. Anyway welcome to the audio forum of audio fora! Instant record holder in reactions to your introduction which is an achievement on itself 🙂
You can debate digital audio/DAC technology big time here for literally hundreds of pages as DIYAudio.com DAC designers have found 40 year old DAC technology to still be in its infancy with many, many gruesome & nerve-racking defects and practically no company nor their design teams (let alone the chip manufacturers) able enough to make stuff right. So no illusions, digital audio is a haunted house with no escape and YOU the poor listener possibly becoming deaf.
You can debate digital audio/DAC technology big time here for literally hundreds of pages as DIYAudio.com DAC designers have found 40 year old DAC technology to still be in its infancy with many, many gruesome & nerve-racking defects and practically no company nor their design teams (let alone the chip manufacturers) able enough to make stuff right. So no illusions, digital audio is a haunted house with no escape and YOU the poor listener possibly becoming deaf.
Last edited:
With file formats, "lossy" vs. "lossless" typically refers to the effects of generational loss. That is, if you take a "lossless" file and resave it to another "lossless" file, you can always reproduce the original file, bit for bit. CD audio is defined as lossless since it can be encoded in other lossless formats (say, WAV -> FLAC -> ALAC -> AIFF) but the original bitperfect WAV can always be extracted. You can verify that with a checksum on the data -- if it varies even by one bit, the checksums will not match.
"Lossy", on the other hand, suffers from generational loss. When you send a file through an MP3 encoder several times, you can never reproduce the original because that data is thrown out. One interesting fact about most modern audio encoders is that they contain a model of the human ear, and one of their primary ways of reducing file size is to (more or less aggressively) discard information that is affected by "masking" which, in this case, is the recovery time of the hairs in your ears (cilia) that are sensitive to a certain frequency. If a second tone occurs after a first one that would be "heard" by the same cilia, then that tone is removed.
Several years ago I had a colleague that was doing his doctoral work on 3D microscopy for vinyl records, using a white light interferometer to scan the grooves, and then constructing a mathematical model of a needle to "play" the 3D models. I learned a lot about the physical nature of LPs and their production by talking with him about his work. Strange things, like the holes in the middle of LPs are not always perfectly on center! Also, that there are physical limitations to the encoding methods of the LP (the "bumps" can only be so big or so small, and depend on the quality of the cutter that cut the physical grooves into the master) that you can only realize when you have a theoretically perfect "needle" to play them. (And with this method you could use image manipulation to "fix" scratches and dust!)
Anyway, LPs are also a lossy format, if by lossy you are defining it as removed information from the original performance, or even a higher resolution medium. Just like in photography: You wouldn't say a photograph taken with a film camera is "lossy" but you can never reproduce all the different refractions of light to bring true depth back to the medium. You also need to factor in equalization and processing: Mixing and mastering necessarily involves changing the audio such that the output is changed and can never be recovered. And LPs are mastered to bump or remove certain frequencies -- this is the whole RIAA curve thing, where an EQ curve is applied to the signal that goes onto the vinyl disc to work around limitations of the physical medium. This is then "decoded" by the inverse RIAA circuitry on the playback equipment. So you could also say that this is a form of compression, since the "original" signal can no longer be accurately reproduced.
And there's also the physicality of the medium -- every time you drag the needle through the groove, you wear down the "signal" a little bit more. So if you have played a record a couple dozen times, can you still say it's "lossless"? Probably not. You've shaved off some of the highs and flattened the bumps for the lows, making both the top and bottom end a bit less defined.
All that is to say that there needs to be some precision about the terms "lossy" and "lossless." Capturing the physical world into any medium, digital or otherwise, is a lossy process. But quality of reproduction does not depend on the presence or absence of data -- as noted, LPs are very "lossy" to the original signal, but many people prefer them. That's OK! It'a a bit like a wood fire: It's nice, comfortable, cozy, and pleasant, but it is not an clean, efficient, or particularly effective method of heating a space.
And for those following from home, here's a little summary of the different numbers, since some of them overlap (and the units are not always clear):
"Lossy", on the other hand, suffers from generational loss. When you send a file through an MP3 encoder several times, you can never reproduce the original because that data is thrown out. One interesting fact about most modern audio encoders is that they contain a model of the human ear, and one of their primary ways of reducing file size is to (more or less aggressively) discard information that is affected by "masking" which, in this case, is the recovery time of the hairs in your ears (cilia) that are sensitive to a certain frequency. If a second tone occurs after a first one that would be "heard" by the same cilia, then that tone is removed.
Several years ago I had a colleague that was doing his doctoral work on 3D microscopy for vinyl records, using a white light interferometer to scan the grooves, and then constructing a mathematical model of a needle to "play" the 3D models. I learned a lot about the physical nature of LPs and their production by talking with him about his work. Strange things, like the holes in the middle of LPs are not always perfectly on center! Also, that there are physical limitations to the encoding methods of the LP (the "bumps" can only be so big or so small, and depend on the quality of the cutter that cut the physical grooves into the master) that you can only realize when you have a theoretically perfect "needle" to play them. (And with this method you could use image manipulation to "fix" scratches and dust!)
Anyway, LPs are also a lossy format, if by lossy you are defining it as removed information from the original performance, or even a higher resolution medium. Just like in photography: You wouldn't say a photograph taken with a film camera is "lossy" but you can never reproduce all the different refractions of light to bring true depth back to the medium. You also need to factor in equalization and processing: Mixing and mastering necessarily involves changing the audio such that the output is changed and can never be recovered. And LPs are mastered to bump or remove certain frequencies -- this is the whole RIAA curve thing, where an EQ curve is applied to the signal that goes onto the vinyl disc to work around limitations of the physical medium. This is then "decoded" by the inverse RIAA circuitry on the playback equipment. So you could also say that this is a form of compression, since the "original" signal can no longer be accurately reproduced.
And there's also the physicality of the medium -- every time you drag the needle through the groove, you wear down the "signal" a little bit more. So if you have played a record a couple dozen times, can you still say it's "lossless"? Probably not. You've shaved off some of the highs and flattened the bumps for the lows, making both the top and bottom end a bit less defined.
All that is to say that there needs to be some precision about the terms "lossy" and "lossless." Capturing the physical world into any medium, digital or otherwise, is a lossy process. But quality of reproduction does not depend on the presence or absence of data -- as noted, LPs are very "lossy" to the original signal, but many people prefer them. That's OK! It'a a bit like a wood fire: It's nice, comfortable, cozy, and pleasant, but it is not an clean, efficient, or particularly effective method of heating a space.
And for those following from home, here's a little summary of the different numbers, since some of them overlap (and the units are not always clear):
- 20Hz - 20kHz -- range of human hearing (the "bandwidth" of the ear). We sample at 44.1kHz since this is the Nyquist frequency for the theoretical top of human hearing + a little extra room. This was chosen for CDs since it was the highest theoretical quality for hearing, while still being processable by the digital equipment (early 1980s) of the time.
- 44.1 / 48 / 96 / 196 / 384 kHz -- sampling rates (the number of times per second a sample of an analogue waveform is captured) for different digital formats.
- 1 / 16 / 24 bit -- the bit depth, or the size of the number that is used to represent each sample. 16 bit (2^16) = ~65,000 possible values for each sample; 24 bit (2^24) = ~16 million possible values. (1-bit DSD operates differently from 16 bit PCM)
- 128 / 192 / 256 / 320 kbps -- the bitrate of a digital files, or the rate at which a decoder needs to be "fed". Higher bitrates are better quality but require more processing power. Also, different format encoders can be more or less efficient at the same bitrate. AAC at 192 kbps is roughly equivalent to MP3 at 256 kbps.
It's worth to mention that most (if not all) bad "digital sound" comes from the source, in the studios, and this is made on purpose by producers.
I have many friends who own studios and they have a strong fetish of using compressors everywhere. I'm talking about of reducing volume variations so as to achieve a more energy dense material or more "compact" sound. And after mixing down, from multi-channel to stereo, there are the mastering phase, where compression goes even further and some times clip the signal on purpose to get extra energy density.
In addition, some producers limit high frequency response from hi-hats, cymbals, splashes etc. I once observed in person a producer applying this limit on drum cymbals. He said: "let's limit to 12kHz for better sounding". Yes, I know, it's crazy! In the extreme world, it's not uncommon to find material called "lo-fi".
It's increadible, but sadly true in popular music - jazz and classical are not much affected.
If you observe the VU metter on mixing consoles, digital recorders etc based on 24bit/96kHz in a modern studio, they are always peaking on yellow or even red leds. So, even with a superb high headroon we have today (24bit -> 144dB), they still push the volume to the limit.
This procedure was needed when we had limited media such as k7 tape of vinyl. If we record well below the maximum, we used to have noise. This is not the case today, even with 16bit/44.1kHz (96dB).
They say that today people listen to music on limited equipment, such as smartphones, small battery operated mini box, while runing, driving, making other stuff and it's better that the sound is limited and compressed. Again, sad, but it's the reality.
Back in vinyl recording, compression was more used to accomodate the original highly dynamic material from the 24-channel 2-inch tape recorders to the 60 to 70dB dynamic range of vinyl (very poor compared to the 96dB of a CD) in order to achieve a reasonable signal to noise ratio.
So compression was used with carefull, only to the need, and this is probably the main reason that vinyl used to sound better. If we could take the original material on tape and record it flat and direct to CD without the traditional "digital remastering" proceess we would have marvelous CD to listen to. I'm refering to pop music.
In my opinion, the bad "digital sound" is not created by the digital media (sampling rate, bit resolution, ADC/DAC etc) but by what producers record, on purpose, in the digital media.
I have many friends who own studios and they have a strong fetish of using compressors everywhere. I'm talking about of reducing volume variations so as to achieve a more energy dense material or more "compact" sound. And after mixing down, from multi-channel to stereo, there are the mastering phase, where compression goes even further and some times clip the signal on purpose to get extra energy density.
In addition, some producers limit high frequency response from hi-hats, cymbals, splashes etc. I once observed in person a producer applying this limit on drum cymbals. He said: "let's limit to 12kHz for better sounding". Yes, I know, it's crazy! In the extreme world, it's not uncommon to find material called "lo-fi".
It's increadible, but sadly true in popular music - jazz and classical are not much affected.
If you observe the VU metter on mixing consoles, digital recorders etc based on 24bit/96kHz in a modern studio, they are always peaking on yellow or even red leds. So, even with a superb high headroon we have today (24bit -> 144dB), they still push the volume to the limit.
This procedure was needed when we had limited media such as k7 tape of vinyl. If we record well below the maximum, we used to have noise. This is not the case today, even with 16bit/44.1kHz (96dB).
They say that today people listen to music on limited equipment, such as smartphones, small battery operated mini box, while runing, driving, making other stuff and it's better that the sound is limited and compressed. Again, sad, but it's the reality.
Back in vinyl recording, compression was more used to accomodate the original highly dynamic material from the 24-channel 2-inch tape recorders to the 60 to 70dB dynamic range of vinyl (very poor compared to the 96dB of a CD) in order to achieve a reasonable signal to noise ratio.
So compression was used with carefull, only to the need, and this is probably the main reason that vinyl used to sound better. If we could take the original material on tape and record it flat and direct to CD without the traditional "digital remastering" proceess we would have marvelous CD to listen to. I'm refering to pop music.
In my opinion, the bad "digital sound" is not created by the digital media (sampling rate, bit resolution, ADC/DAC etc) but by what producers record, on purpose, in the digital media.
Hi I knew one with a small studio recording local artists so that caught my interest. To my extreme surprise masters were delivered in 320 kbit MP3 and converted again. About nothing stays unmanipulated either. For the HiFi minority a cold shower really.
We can all give a crazy amount of time to minute details and (often extremely exaggerated) technical deficiencies but nothing beats reality. Also most material is recorded so so which makes it even less intelligent to put so much time in technical absolute perfection IMHO.
We can all give a crazy amount of time to minute details and (often extremely exaggerated) technical deficiencies but nothing beats reality. Also most material is recorded so so which makes it even less intelligent to put so much time in technical absolute perfection IMHO.
Last edited:
OP was looking for a Phono to USB adapter. While DACs used for playback may have some impact on the final sound the ADC used in phono-to-USB conversion has much more impact. Differences between ADCs are much more audible than between DACs. Discussion of DACs in this context is more or less pointless as the milk has already been spilled.
You are right, the detail that OP is interested in phono to USB conversion was also buried in the first post. I count the word "CD" 5 times though and more than 70% of the first post rant was about bandwidth/frequency range confusion and CD playback versus superior vinyl playback (why the superior vinyl must be converted to hires digital is unclear).
Last edited:
Off topic: the song "I Hate You!" by Kirk Thatcher and The Edge of Etiquette was deliberately recorded with the worst microphones they could find, in the hall outside a studio. It was recorded specifically for the movie Star Trek IV and it had to sound like obscure 1980's punk rock.