|
|
Joined: May 2000
Posts: 8,364 Likes: 14
President of the Imperial Galactic Government
|
|
President of the Imperial Galactic Government
Joined: May 2000
Posts: 8,364 Likes: 14 |
Hi Evdberg,
I'm not sure I agree. There's a disagreement above on what bits may/may not be being manipulated. HOW does this impact the sound (if it does) in the way we might perceive?
In other words, if we are talking about 23bit vs 24bit, then it seems to be a fairly significant impact on the presentation of the sound. Is that possible ?
Put another way around (and with respect too) I never found your 24bit conversion program from 16bit ever made this type of difference ? The logic cannot be so remote ?
thoughts ? Read my post further back. 23 vs 24 bits should be insignificant if there was such thing as a perfect DAC. There is no such thing. That's what folks are confusing. I was confused about this too and now it makes sense. In reality we are talking on the order of 17bits vs 18bits or 18 vs 19. You can argue where it is exactly, but it is NOT 23 vs 24. If a perfect DAC existed we'd all have one and there would be no argument over differences in who makes a better DAC. 24 bits exists mostly for headroom in processing, mixing etc. If I change my speedometer from 120mph limit too 200mph limit will my car now go faster than it did. Will it handle better now at 120mph beacause it's not running near the speedometer's limit. This is not a perfect analogy but it's not that far off. Don't assume just because it's labeled 24bit it can actually deliver 24bit to your ears.
- Fiddler (IMHO)
If you think listening to music is fun, try learning an instrument. HT: G61RSL, 4x5K, 5.5KHC, 2xSC6000, Transporter, HD621, Philips BDP7501 4K, Tivo, 65" LG OLED 4K C6 Computer: Foobar2000, 5K, Mojo, HiFiMan HE1000 V2
|
|
|
|
|
Joined: Dec 2000
Posts: 866 Likes: 2
Pan-dimensional being
|
|
Pan-dimensional being
Joined: Dec 2000
Posts: 866 Likes: 2 |
Put another way around (and with respect too) I never found your 24bit conversion program from 16bit ever made this type of difference ? Personally I hear a lot of difference with AR, even George did hear it and he claims he is deaf! :-) However, with current knowledge it seems that the 3dB headroom AR uses may lower the quality and since the results are not the same with every album, this could mean that some albums can sound worse with AR.
|
|
|
|
|
Joined: May 2000
Posts: 8,364 Likes: 14
President of the Imperial Galactic Government
|
|
President of the Imperial Galactic Government
Joined: May 2000
Posts: 8,364 Likes: 14 |
Here is another test (one of dozens that keep popping into my head).
Take a 16bit piece of music in a wave file.
Convert it to 24bit.
Now stick garbage in the lower 8bits (maybe a nasty modest freq square wave (say a buzz as 440hz)).
Now play it. Bet you'll hardly hear the garbage. Because it's SO insignificant.
Let's say you do hear the garbage.
Now repeat the process and stick garbage in the lower 7bits.
Do you hear that?
Repeat doing less and less bits until you don't hear the garbage.
That's how "Effectively" how many bits you are HEARING.
And I guarantee you will stop hearing anything, long before you reach putting garbage in the least significant bit.
The offset Meridian applies pushes what is audible down into inaudible by a MINIMUM of 3dB.
- Fiddler (IMHO)
If you think listening to music is fun, try learning an instrument. HT: G61RSL, 4x5K, 5.5KHC, 2xSC6000, Transporter, HD621, Philips BDP7501 4K, Tivo, 65" LG OLED 4K C6 Computer: Foobar2000, 5K, Mojo, HiFiMan HE1000 V2
|
|
|
|
|
Joined: May 2000
Posts: 8,364 Likes: 14
President of the Imperial Galactic Government
|
|
President of the Imperial Galactic Government
Joined: May 2000
Posts: 8,364 Likes: 14 |
Another test.
Take the test above and find the point where you last heard a change.
Now calibrate system 6dB lower.
Poof now you can't hear that "garbage" because you pushed it out of your DAC (and ears) audible range.
Now what if that garbage was "real" music. You just pushed out 6dB of the good stuff.
This is not anything new. We also knew digital attenuation is bad. But we fooled ourselves into thinking it's ok in 24bit. But that assumed a perfect 24bit DAC.
- Fiddler (IMHO)
If you think listening to music is fun, try learning an instrument. HT: G61RSL, 4x5K, 5.5KHC, 2xSC6000, Transporter, HD621, Philips BDP7501 4K, Tivo, 65" LG OLED 4K C6 Computer: Foobar2000, 5K, Mojo, HiFiMan HE1000 V2
|
|
|
|
|
Joined: May 2000
Posts: 8,364 Likes: 14
President of the Imperial Galactic Government
|
|
President of the Imperial Galactic Government
Joined: May 2000
Posts: 8,364 Likes: 14 |
Another test.
Hook SqueezeBox/Transporter direct to DSP Speaker.
Take a good 16bit sample and listen at 63 (on DSP Speaker Display)
Take your SqueezeBox/Transporter out of Max Volume Mode.
Lower volume by 36dB (6 bits)
Raise Volume on DSP Speaker to 99dB (+36dB).
It should still be getting all 16bits.
Bet you it sounds like crap (it should be no louder or softer). Because now your operating it the worst range of the DAC.
You could do the same test by using ReplayGain, or editing with Audacity.
- Fiddler (IMHO)
If you think listening to music is fun, try learning an instrument. HT: G61RSL, 4x5K, 5.5KHC, 2xSC6000, Transporter, HD621, Philips BDP7501 4K, Tivo, 65" LG OLED 4K C6 Computer: Foobar2000, 5K, Mojo, HiFiMan HE1000 V2
|
|
|
|
|
Joined: May 2000
Posts: 8,364 Likes: 14
President of the Imperial Galactic Government
|
|
President of the Imperial Galactic Government
Joined: May 2000
Posts: 8,364 Likes: 14 |
Another test.
Do the test I said a couple posts back inserting garbage in lower and lower bits.
But do it against a "Silent" WAV file (all zeros). No need for music to clutter up the test.
- Fiddler (IMHO)
If you think listening to music is fun, try learning an instrument. HT: G61RSL, 4x5K, 5.5KHC, 2xSC6000, Transporter, HD621, Philips BDP7501 4K, Tivo, 65" LG OLED 4K C6 Computer: Foobar2000, 5K, Mojo, HiFiMan HE1000 V2
|
|
|
|
|
Joined: Sep 2007
Posts: 1,980 Likes: 55
Knows where his towel is
|
|
Knows where his towel is
Joined: Sep 2007
Posts: 1,980 Likes: 55 |
Take the test above and find the point where you last heard a change.
Now calibrate system 6dB lower.
Poof now you can't hear that "garbage" because you pushed it out of your DAC (and ears) audible range.
Now what if that garbage was "real" music. You just pushed out 6dB of the good stuff Nope, you just got rid of the bad stuff. If you filled the 8 LSB with nasty noise in a 24 bit signal, it would be at -96dB. If you can hear that signal, with other musical content, then you truly have 'golden' ears. I'd even go as far to say you wouldn't hear the signal if no other content was present. (played back at the same level as if a signal was present, no cheating and turning it up!) Steve.
|
|
|
|
|
Joined: Sep 2007
Posts: 1,980 Likes: 55
Knows where his towel is
|
|
Knows where his towel is
Joined: Sep 2007
Posts: 1,980 Likes: 55 |
Another test.
Hook SqueezeBox/Transporter direct to DSP Speaker.
Take a good 16bit sample and listen at 63 (on DSP Speaker Display)
Take your SqueezeBox/Transporter out of Max Volume Mode.
Lower volume by 36dB (6 bits)
Raise Volume on DSP Speaker to 99dB (+36dB).
It should still be getting all 16bits.
Bet you it sounds like crap (it should be no louder or softer). Because now your operating it the worst range of the DAC.
You could do the same test by using ReplayGain, or editing with Audacity. This is exactly the reason for higher bit DACs, better linearity and less noise (circuit and dither). Steve.
|
|
|
|
|
Joined: Sep 2007
Posts: 1,980 Likes: 55
Knows where his towel is
|
|
Knows where his towel is
Joined: Sep 2007
Posts: 1,980 Likes: 55 |
Another test.
Do the test I said a couple posts back inserting garbage in lower and lower bits.
But do it against a "Silent" wav file (all zeros). No need for music to clutter up the test. Do a similar test, start wih the signal at 0dBFS, keep reducing the signal by 6dB until you can't hear it at all, this will show you the absolute resolution capability of your whole system, including your room (ambient noise) and your ears! I bet nobody gets anything near 24bits! Steve.
|
|
|
|
|
Joined: May 2000
Posts: 8,364 Likes: 14
President of the Imperial Galactic Government
|
|
President of the Imperial Galactic Government
Joined: May 2000
Posts: 8,364 Likes: 14 |
Another test.
Do the test I said a couple posts back inserting garbage in lower and lower bits.
But do it against a "Silent" wav file (all zeros). No need for music to clutter up the test. Do a similar test, start wih the signal at 0dBFS, keep reducing the signal by 6dB until you can't hear it at all, this will show you the absolute resolution capability of your whole system, including your room (ambient noise) and your ears! I bet nobody gets anything near 24bits! Steve. Exactly, silly me making it much harder than it needs to be. You may not even hear anything in the low 8 bits. Actually now that you say that you could just use that test signal I made in the wicky (full scale 440hz) and keep lowering it using Digital Volume on the Squeezebox. I'm not exactly sure the scale/step size though on the SqueezeBox/Transporter though.
- Fiddler (IMHO)
If you think listening to music is fun, try learning an instrument. HT: G61RSL, 4x5K, 5.5KHC, 2xSC6000, Transporter, HD621, Philips BDP7501 4K, Tivo, 65" LG OLED 4K C6 Computer: Foobar2000, 5K, Mojo, HiFiMan HE1000 V2
|
|
|
|
|
Joined: May 2000
Posts: 8,364 Likes: 14
President of the Imperial Galactic Government
|
|
President of the Imperial Galactic Government
Joined: May 2000
Posts: 8,364 Likes: 14 |
Nope, you just got rid of the bad stuff.
If you filled the 8 LSB with nasty noise in a 24 bit signal, it would be at -96dB. If you can hear that signal, with other musical content, then you truly have 'golden' ears.
I'd even go as far to say you wouldn't hear the signal if no other content was present. (played back at the same level as if a signal was present, no cheating and turning it up!)
Steve. Agreed.
- Fiddler (IMHO)
If you think listening to music is fun, try learning an instrument. HT: G61RSL, 4x5K, 5.5KHC, 2xSC6000, Transporter, HD621, Philips BDP7501 4K, Tivo, 65" LG OLED 4K C6 Computer: Foobar2000, 5K, Mojo, HiFiMan HE1000 V2
|
|
|
|
|
Joined: May 2000
Posts: 8,364 Likes: 14
President of the Imperial Galactic Government
|
|
President of the Imperial Galactic Government
Joined: May 2000
Posts: 8,364 Likes: 14 |
By the way if folks are still following the back and forth between Steve and I and if you wanted to any these tests (or understand them) then read on.
It is FAIR game to turn up the volume on the DSP Speaker to hear these low level bits. That's the thrust of the whole topic.
How much can you attenuate digitally and still bring it back audibly in analog amplification.
This is IMPORTANT to understand (in the context of the 3dB tweak).
So let's say you take a test signal (like the FLAC I used to do +3dB testing).
Choose a modestly low volume on DSP speaker, say 40 (this will give us 60dB of analog headroom).
Now lower it on SqueezeBox until you just barely hear it (probably on the order of -10dB). You could even use a meter !!
Now lower it 3dB more on the squeezebox and raise the DSP 3dB. You want to lower more than 3dB and then raise DSP by same amount. So it clearly goes away in digital and amplifies back in Analog.
Now your listening to the DAC resolve better than your ears or your room can (And there by taking those out of the equation).
Now lower it 3dB more on SqueezeBox and raise DSP speaker 3dB.
Keep going until you can't get the signal back by raising the DSP speaker 3dB.
Now you are at the resolving limit of the DAC (Approximately).
You could do the test above in any increment it will just effect your precision. 6dB may be easier.
- Fiddler (IMHO)
If you think listening to music is fun, try learning an instrument. HT: G61RSL, 4x5K, 5.5KHC, 2xSC6000, Transporter, HD621, Philips BDP7501 4K, Tivo, 65" LG OLED 4K C6 Computer: Foobar2000, 5K, Mojo, HiFiMan HE1000 V2
|
|
|
|
0 members (),
169
guests, and
2
robots. |
|
Key:
Admin,
Global Mod,
Mod
|
|
|
Forums18
Topics31,151
Posts296,947
Members5,503
| |
Most Online4,788 May 22nd, 2026
|
|
|
|