HD Digital Audio is BS! Great (albeit long) Video
- Thread starter Flint
- Start date
Ahhhh….. that's the curiosity isn't it!?!?!
I've rambled on about it in the past, but I'll say it again. You will not find a piece of music available on the various formats (LP, CD-Audio, MP3, HD Digital, etc.) which hasn't been remastered for each format. I've even discovered some downloadable MP3 versions of song from CDs I own don't have the same frequency balance and mix. In other words, it is damn near impossible to directly compare the various commercial audio formats because the content in the formats is different sounding BEFORE it was put on the disc or disk or music store. Now, in most cases the CD and MP3 versions are the same original content, but I've never heard a HD version of any song or recording which was remotely the same mix or master as the CD version.
Secondly, as alluded to in the video, IMD can easily be made worse by having enough ultra-sonic information in a recording. In other words, a bunch of inaudible content in the frequency range above 20,000 can generate intermodulation distortion in frequencies we can easily hear when amplified by transistor amps. So, if a recording does have a lots of ultrasonic content, it can harm the lower frequencies we can earn.
Thirdly, and this is what I believe I most often the cause of people's preference for HD audio, our brains' abilities to understand anything we hear is shockingly impacted by our other knowledge of the situation. If you believe a high resolution audio file should sound better because it just seems to make sense, and if you believe a high resolution audio file is playing at the time, you will believe it sounds better - even if neither of mitigating beliefs are true. You could be listening to an MP3 at 320kbps and the person who told you that HD audio must be better could have been lying, but you'll never know and you'll convince yourself unconsciously that what you are hearing is the best you've ever heard. Since it is unconscious, it really doesn't matter - as long as you are enjoying it. The reason to care is to save money and time and effort on buying content you already own just to get something you think sounds better because, you know, Stereophile says so.
Razz
Well-Known Member
This almost comes off as a conspiracy theorist. But now I'm starting to wonder....
Do you think artist record a separate "cleaned up" version of a song just to sell it as HD???? I guest that is possible.
Do I think they do it? YES!
They don't re-record the music, they re-mix or re-master it where they can apply more or less compression, adjust the EQs differently, and even change the levels of the various musical tracks. With modern digital recording, today's recordings usually have a different audio file for each microphone feed used in the music. That allows them the adjust with impunity any part, even slightly increasing the bass drum, re-equalizing the bass guitar, or in rare cases use a different take of a part, if they want.
Why do it? Because the people promoting HD digital audio desperately need the public to think it sounds better in order to ensure the public is willing to pay more for it. So, much like DTS does with soundtracks, HD Audio companies often have specific requests for the content produced for HD music sales. They often want a little more air in the treble, less compression overall, and more deep bass. They often also ask for slightly less midrange - even a difference of 0.5dB between 800Hz and 5kHz will make a recording sound smoother and less "honky".
Yes, they do it and I know they do it, as does everyone in the music industry.
We've known for years that LP mixes are different from CD and MP3 mixes. Why we don't readily accept it for HD Digital audio is strange to me.
I will add, the marketing department for HD audio labels will tell you they are remixing the music to take full advantage of the higher resolution format without the limitations of CD audio. Of course, that's all 100% bullshit, but consumers fall for it.
As shown in the video, CD-Audio can extremely accurately reproduce 95% of everything our brains can hear.
When the DVD-Audio format was being developed and standardized, the researchers who did the basic work on what the format should accomplish stated their researching in a series of technical white papers, also published in Audio Magazine. The most ideal format which solved absolutely every limitation of the CD-Audio format (16 bit, 44,100Hz sampling rate per channel) was a format which was 18 bit and 64kHz sampling rate per channel.
The reason for 18 bit sample depth was that analog components such as resistors, capacitors, rectifiers, etc. all had a self-noise which prevented any audio component to have a S/N effectively greater than 104dB. So, with 18 bits and using dithering, the digital signal would exceed the limits of the equipment reproducing it by more than 3dB.
The reason for 64kHz sample rate wasn't to reproduce frequencies above 22kHz (the limit of a decent 48kHz system already in use by DAT, MD, and other common digital formats). Instead, the reason for the higher sampling rate was to eliminate the ringing and phase issues with the analog filter put on the DAC to remove the sampling rate noise. At the time (mid-90s), the biggest difference between consumer grade DACs was how it's low pass filter performed, and if you recall the quality test bench reviews in the magazines back then, that was measured in impulse response (preringing and postringing), phase, and micro-amplitude response characteristics. By raising the sampling rate to 64kHz, the low pass filter could be shifted up to 30kHz rather than 20Khz or 22kHz necessary for 44.1k and 48k formats, and thus effectively eliminate the sometimes audible negative attributes of consumer grade DACs. Remember, back then issues like RF/EMI noise and Jitter were still extremely common and often audible.
However, since those days DACs have been refined to the point that we no longer talk about those issues which were audibly in some cases in consumer gear.
The reason DVD-Audio settled on 24bit / 96kHz was that the pro market was already flooded with hardware which easily supported those specs and it would be affordable to design, manufacture, and deliver the gear. They knew it was overkill, but it was easiest to deliver and market. However, the Marketing Departments all went ape-shit and sold us on the idea that higher numbers ALWAYS translate to better sound. That tested well with consumers, so 192kHz sampling was added to the format standard shortly after that and now we have some people promoting 384kHz sampling rates, which is outright stupid.
As shown in the video, CD-Audio can extremely accurately reproduce 95% of everything our brains can hear.
When the DVD-Audio format was being developed and standardized, the researchers who did the basic work on what the format should accomplish stated their researching in a series of technical white papers, also published in Audio Magazine. The most ideal format which solved absolutely every limitation of the CD-Audio format (16 bit, 44,100Hz sampling rate per channel) was a format which was 18 bit and 64kHz sampling rate per channel.
The reason for 18 bit sample depth was that analog components such as resistors, capacitors, rectifiers, etc. all had a self-noise which prevented any audio component to have a S/N effectively greater than 104dB. So, with 18 bits and using dithering, the digital signal would exceed the limits of the equipment reproducing it by more than 3dB.
The reason for 64kHz sample rate wasn't to reproduce frequencies above 22kHz (the limit of a decent 48kHz system already in use by DAT, MD, and other common digital formats). Instead, the reason for the higher sampling rate was to eliminate the ringing and phase issues with the analog filter put on the DAC to remove the sampling rate noise. At the time (mid-90s), the biggest difference between consumer grade DACs was how it's low pass filter performed, and if you recall the quality test bench reviews in the magazines back then, that was measured in impulse response (preringing and postringing), phase, and micro-amplitude response characteristics. By raising the sampling rate to 64kHz, the low pass filter could be shifted up to 30kHz rather than 20Khz or 22kHz necessary for 44.1k and 48k formats, and thus effectively eliminate the sometimes audible negative attributes of consumer grade DACs. Remember, back then issues like RF/EMI noise and Jitter were still extremely common and often audible.
However, since those days DACs have been refined to the point that we no longer talk about those issues which were audibly in some cases in consumer gear.
The reason DVD-Audio settled on 24bit / 96kHz was that the pro market was already flooded with hardware which easily supported those specs and it would be affordable to design, manufacture, and deliver the gear. They knew it was overkill, but it was easiest to deliver and market. However, the Marketing Departments all went ape-shit and sold us on the idea that higher numbers ALWAYS translate to better sound. That tested well with consumers, so 192kHz sampling was added to the format standard shortly after that and now we have some people promoting 384kHz sampling rates, which is outright stupid.
Razz
Well-Known Member
Yeah, seeing how it works in the example video, I understand how they may want to use 48kHz or in your sited version 64kHz as they are being used for making things work better. But who came up with 96kHz and what is the purpose? 192? 384? WTF??
I never really knew how sampling rates worked but now knowing, it almost pisses me off.
I never really knew how sampling rates worked but now knowing, it almost pisses me off.
When designing digital circuits, they had quartz crystals for timing commonly used in communications and computing platforms which easily got you 48kHz with multipliers. So, 48kHz was easier to achieve than 44.1kHz in terms of designing clocking circuits. Then when any sampling rate higher than 48kHz was desired, it was WAY easier to use a multiple of 48kHz than anything else, thus the whole 96kHz, 192kHz, and 384kHz. So, it saved a ton of money and effort to design a 96kHz recording hardware platform than anything between 48 and 96.
Yeah.... I think most people on this forum have either already discovered this or just don't want to know.
Just the same, I am glad someone did a good job of making the point very clear. The entire HD Audio market is frustrating to me because some of the mixes I've heard in HD are great and I cannot get them in a format I want which is easy to enjoy on my phone and such.
Just the same, I am glad someone did a good job of making the point very clear. The entire HD Audio market is frustrating to me because some of the mixes I've heard in HD are great and I cannot get them in a format I want which is easy to enjoy on my phone and such.
BUMP!
I've noticed lately that lossy compressed MP3 has gotten significantly better than it was even 5 years ago when I started relying on it more and more. Back in the early 2000s MP3 was terrible and even with the expensive state of the art encoders from companies like Meridian and the Fraunhofer Institute still required data rates no less than 192kbps and ideally about 320kbps to sound remotely acceptable in a high grade system with good music. Around 2010 it got better as some of the patents on encoding tech expired and free encoders suddenly got access to the commercial grade performance and three to four years after that some significant advancements were introduced to make it pretty decent at even lower data rates.
Meanwhile the hobbyist community shifted to FLAC for serious music files and lossy compression fell out of favor, but with Bluetooth audio needing improvement companies like Qualcomm created their APT-X compression using state of the art realtime encoding which is extremely close to ideal, but still shy of FLAC.
So, today I am perfectly happy giving up on my vast library of WMA encoded music (in the day good quality at decent cost, or free, required using Microsoft's WMA or Apple's AAC to get good results always better than free MP3). Streaming MP3 music is more than good enough for anything shy of serious critical listening or auditioning speakers and sound systems.
Still, we DO NOT NEED high resolution uncompressed digital audio to get the most out of our sound systems and enjoy our beloved music.
CD Audio is still a better format than 80% of the available content can utilize. For the 20% of music created which could benefit from more resolution, most reproductions systems cannot reproduce the added detail or dynamics, and few us would never ever notice the difference even if we had one of those systems. Our auditory limitations cannot realistically handle more than what an 18-bit 44.1kHz PCM system can reproduce - especially with simple dither applied to the tracks.
So, yes... a 16 bit format is slightly limited if the music content is actually using the extra 2 bits of dynamic range, like a really damn good classical recording or a nature recording. More than 18-bit is pointless in terms or reproduction.
I've noticed lately that lossy compressed MP3 has gotten significantly better than it was even 5 years ago when I started relying on it more and more. Back in the early 2000s MP3 was terrible and even with the expensive state of the art encoders from companies like Meridian and the Fraunhofer Institute still required data rates no less than 192kbps and ideally about 320kbps to sound remotely acceptable in a high grade system with good music. Around 2010 it got better as some of the patents on encoding tech expired and free encoders suddenly got access to the commercial grade performance and three to four years after that some significant advancements were introduced to make it pretty decent at even lower data rates.
Meanwhile the hobbyist community shifted to FLAC for serious music files and lossy compression fell out of favor, but with Bluetooth audio needing improvement companies like Qualcomm created their APT-X compression using state of the art realtime encoding which is extremely close to ideal, but still shy of FLAC.
So, today I am perfectly happy giving up on my vast library of WMA encoded music (in the day good quality at decent cost, or free, required using Microsoft's WMA or Apple's AAC to get good results always better than free MP3). Streaming MP3 music is more than good enough for anything shy of serious critical listening or auditioning speakers and sound systems.
Still, we DO NOT NEED high resolution uncompressed digital audio to get the most out of our sound systems and enjoy our beloved music.
CD Audio is still a better format than 80% of the available content can utilize. For the 20% of music created which could benefit from more resolution, most reproductions systems cannot reproduce the added detail or dynamics, and few us would never ever notice the difference even if we had one of those systems. Our auditory limitations cannot realistically handle more than what an 18-bit 44.1kHz PCM system can reproduce - especially with simple dither applied to the tracks.
So, yes... a 16 bit format is slightly limited if the music content is actually using the extra 2 bits of dynamic range, like a really damn good classical recording or a nature recording. More than 18-bit is pointless in terms or reproduction.
walls
Well-Known Member
Yeah I got around 10 minutes in and wanted to push my eyes in. Lol
Oh, I didnt get that far, but i was at work and only had about 10 minutes to good off so I didn't think it was worth starting.
walls
Well-Known Member
I don’t know about all the measurements and stuff, honestly I like what I like and if someone’s graph or chart tells me that it’s not perfectly flat or that the bass is bloated I don’t care. I eq quote a bit of what I listen to, for example AC/DCs first couple albums are so tinny sounding that if I don’t cut back the treble and boost the mods and bass a bit I don’t enjoy them. And that’s what it’s about anyways right?
malsackj
Well-Known Member
Yes and the m-audio Delta 44 or 66 allows me to capture 4 channels each. The X32 mixers allow for 32 channels to be captured. Then like Flint has said it allows me to playback and mix it down to different versions. Some of the differences will come not from the mixer. Some of the differences that is noted will be the difference in various mic's and the preamps in front of the AD conversion. These parts can have some effect on what is recorded because the analog is in front of the digital conversion. Better Mics and preamps can help.
malsackj
Well-Known Member
I also feel that this is because they cannot sell the DAW mixing software and the hardware to support the interface. Like computers. You need faster and updated computers to run the software and applications. Same is true in the Digital video and audio worlds. Dante is a new product in the world of audio and video. This product is different because it works going across the switched based networks that most computers are working on.