Dsd?

Jan 8, 2019 at 8:07 PM Post #31 of 56
Thanks for that explanation. I can understand that boosting the signal just before reaching that moment, and probably most relevant in the mastering domain, but how does that work with a more general pre-emphasis, eg a linear boost/cut on some early CDs or the RIAA curve on LP playback (or Dolby NR for that matter)? For example, if we boost all the higher frequencies to deal with tape hiss, why wouldn't the hiss also be boosted and remain as is when reversed?
because we're talking about the hiss generated on playback specifically. whatever noise came into the recording will be there to stay no matter what. once it's in the signal, there is not much to do about it(maybe gating if the noise is low enough compared to music).
the idea of a pre-emphasis seems to be as I explained, you get the signal at a frequency much louder than it is, so that when your playback system generates noise at those frequencies, the music will hopefully be well above it in the resulting signal. and we de-emphasize after that noise crept in, but before the output, so the sound is back to it's original signature, but the noise is XdB below what it would have been if we had done nothing.
am I making any sense at all? ^_^
 
Last edited:
Jan 9, 2019 at 3:00 AM Post #33 of 56
16 is great, fantastic! belissimo! however, i do notice a slight change from 16 to 24. mostly to do with volume sensitivity. i've noticed personally that songs in 24bit require more volume and can get louder without sounding scratchy. but thats just me and my setup and my ears. i could be wrong.

What you are hearing could well be correct but your reason for WHY you're hearing it, is not. With digital audio, the maximum peak level (0dBFS) is always the same, regardless of whether the digital audio is 8bit, 16bit, 32bit or whatever. So there cannot be any difference of "volume sensitivity" between 16bit and 24bit. However, there most certainly can be (and sometimes is) a difference in "volume sensitivity" between different masters. What you are describing is exactly what would be expected when comparing two different masters, specifically; two masters that are identical apart from one of them having more audio (dynamic range) compression. The master with less compression would appear to be quieter and would therefore require more volume, while turning-up the level of the compressed master would make the compression artefacts more noticeable and, depending on the playback equipment, may introduce more distortion. Furthermore, we know this scenario (of two different masters, the 16bit version being more compressed) exists because we can measure and compare masters and, some record labels have admitted to doing exactly this (Linn Records for example). Your reason WHY you are hearing a difference is incorrect because if you use the same master for both the 16bit and 24bit releases there will not be any difference in "volume sensitivity" (or scratchiness). Of course, this is at least part of the reason why some releases have more audio compression on the 16bit version; to demonstrate a difference between 16bit and 24bit and thereby mislead audiophiles into believing there is an audible benefit (of 24bit) which justifies a higher price.

For example, if we boost all the higher frequencies to deal with tape hiss, why wouldn't the hiss also be boosted and remain as is when reversed?

It would be. Castleofargh's explanation is essentially correct. Every electrical circuit adds noise (at least thermal noise) and analogue storage media (vinyl or tape) adds considerably more noise. The pre-emphasis (boost) occurs at the end of the mastering process, so while the all the noise of capture, storage, mixing and mastering is all boosted as well, when it's reproduced by the consumer, all the additional noise of the analogue distribution media and the consumer's playback device (turntable or cassette player) is reduced by the amount of boost added at the end of mastering. In the case of early redbook, there is no additional noise from the distribution media (CD) but there was expected to be additional noise from the consumer's CD player, caused by the analogue dithering quantiser and very steep reconstruction filters, and that would have been reduced by the use of "emphasis".

G
 
Jan 9, 2019 at 5:06 AM Post #34 of 56
With regard to pre-emphasis generally, I've never got my head around why boosting the frequencies on recording and then reversing the boost on playback results in noise reduction.
It's actually about using the dynamic range more effectively. Say you have the noise floor at level -100 dB. If you record a signal at level -20 dB without any pre-emphasis, you end up with 80 dB dynamic range. If you do +10 dB pre-emphasis, your signal is at -10 dB and if you get 90 dB dynamic range. If you have a lot of noise in the original signal this doesn't help, but if the original noise is very low, it helps.

Low frequencies tend to set the recording level while higher frequencies are quieter. Pre-emphasis makes the high frequencies "as loud" as lower frequencies optimizing dynamic range usage.
 
Last edited:
Jan 9, 2019 at 5:13 AM Post #35 of 56
For example, if we boost all the higher frequencies to deal with tape hiss, why wouldn't the hiss also be boosted and remain as is when reversed?
We boost the signal which hopefully contains much less noise than tape hiss. The tape doesn't even know the signal was boosted. It just sees strong signal and introduces the same tape hiss it introduces to all signals from weak to strong.
 
Jan 9, 2019 at 5:24 AM Post #36 of 56
I've exactly the same problem, volume with flac 24 drops down
I believe its because theres more information in the file and the studio or recording source didnt make the song louder or have an elevated sound floor. I find the lower the quality the sample rate the louder the song is.. :(
 
Jan 9, 2019 at 5:27 AM Post #37 of 56
I believe its because theres more information in the file and the studio or recording source didnt make the song louder or have an elevated sound floor. I find the lower the quality the sample rate the louder the song is.. :frowning2:
And to go further when I listened few DSD files the volume drop was even bigger
 
Jan 9, 2019 at 5:33 AM Post #38 of 56
And to go further when I listened few DSD files the volume drop was even bigger
I think gregorio makes a better explanation. (Scroll up lol)
I find that listening to two songs from the same artist and studio recording (check cd case and source online) the 24bit has more information and requires more volume because its less compressed.

Now, as for dsd, dxd, etc... i find thoes formats are more for preserving the recording for later release. Archiving and the like, but also for working on the recording itself for release. I personally dont know anyone who listens to dsd or has a collection in that format. Most artists will never release in that format. And if they did, it would be huge name artists that could afford to release in that format. Example, Michael jackson, acdc, tom Jones, abba, rolling stones, etc...
 
Jan 9, 2019 at 6:10 AM Post #39 of 56
I think gregorio makes a better explanation. (Scroll up lol)
I find that listening to two songs from the same artist and studio recording (check cd case and source online) the 24bit has more information and requires more volume because its less compressed.

Now, as for dsd, dxd, etc... i find thoes formats are more for preserving the recording for later release. Archiving and the like, but also for working on the recording itself for release. I personally dont know anyone who listens to dsd or has a collection in that format. Most artists will never release in that format. And if they did, it would be huge name artists that could afford to release in that format. Example, Michael jackson, acdc, tom Jones, abba, rolling stones, etc...
Yes and it's not worth it too considering that a single album can be over 1 Gb but i only wanted to try and it's really not a good deal
 
Jan 9, 2019 at 6:18 AM Post #40 of 56
Yes and it's not worth it too considering that a single album can be over 1 Gb but i only wanted to try and it's really not a good deal
Agreed, don't get suckered into products either that brag about having native DSD and all that DSD w/e. you're going to be burning extra money for a format thats impractical and you'll never use.
Stick to 16 or 24bit. its better for your wallet :P and still sounds amazing.
 
Jan 9, 2019 at 7:50 AM Post #41 of 56
Stick to 16

You should have stopped there IMO. 24 bit is impractical and is bad for your Wallet. Marketing will always attempt to get us to buy something we don't need or find a way to sell us something we already have in a 'new' way and charge us more for the privilege. From my own experience the only 24 bit Albums that have sounded different to me have obviously been remastered or are just louder.

DSD (in this context) is just the next step in reselling. There'll inevitably be more unless people stop buying into it but Marketing is a powerful force and some people are more susceptible to it.
 
Last edited:
Jan 9, 2019 at 7:55 AM Post #42 of 56
You should have stopped there IMO. 24 bit is impractical and is bad for your Wallet. Marketing will always attempt to get us to buy something we don't need or find a way to sell us something we already have in a 'new' way and charge us more for the privilege. From my own experience the only 24 bit Albums that have sounded different to me have obviously been remastered or are just louder.

DSD (in this context) is just the next step in reselling. There'll inevitably be more unless we stop buying buying into it but Marketing is a powerful force.
24bit audio albums in stores near me arent that much more than a standard album. for example a regular CD is around 12-15$ a 24bit is around 18 - 24$ depending on the artist and album released. its not a huge difference. it really depends on what your budget is in music. i've seen 24bit albums go on sale for like 12-14$ also
 
Jan 9, 2019 at 8:04 AM Post #43 of 56
Yes, but it's still reselling it and it's still 'more Money'. You don't need 24 Bit, even if it was the same price, you don't need it. I'd rather save myself the Money and storage space and go with 16 bit, that's all I need, it's all anyone needs but people believe in higher numbers in Audio way too much. YMMV.

(Apologies I was still editting my previous post, seems my Brain woke up slow today)

EDIT ( I did warn you lol)

I was searching for evidence that 24 bit was better a while back after it failed to WOW me and found this thread (It was the thread that brought me into Sound Science) it's quite the read :-

https://www.head-fi.org/threads/24bit-vs-16bit-the-myth-exploded.415361/page-340
 
Last edited:
Jan 9, 2019 at 8:36 AM Post #44 of 56
I think gregorio makes a better explanation. (Scroll up lol)
I find that listening to two songs from the same artist and studio recording (check cd case and source online) the 24bit has more information and requires more volume because its less compressed.

Now, as for dsd, dxd, etc... i find thoes formats are more for preserving the recording for later release. Archiving and the like, but also for working on the recording itself for release. I personally dont know anyone who listens to dsd or has a collection in that format. Most artists will never release in that format. And if they did, it would be huge name artists that could afford to release in that format. Example, Michael jackson, acdc, tom Jones, abba, rolling stones, etc...
'
We obviously compared different CDs and 24bit files of the same artist/album.

Most of my CDs were issued in the 1980s/1990s and they generally are much quieter and more dynamic (less compressed) than most of my 24 bit music issued after 2010 (compare any hi res ELO album to the original CD issues as just one example of hundreds). This has nothing to do with 16 or 24 bits, it is more a reflection of mastering trends with regard to loudness and compression over the decades.

I'm also unsure what you mean by the 24bit music files having more information when a recording doesn't even have the dynamic range a CD is capable of - in other words, how does having more zeros benefit the listening experience?
 
Last edited:

Users who are viewing this thread

Back
Top