Are you ever going to hear a difference? Say they're both 44.1Hz - can anyone hear the difference the extra bits make?
No. It’s not the bits. It’s the filter and how it pushes noise out further. and yes the difference is noticeable. Big time
Yes, that’s what makes hi-rez files, hi-rez. The difference is very noticeable. Although true redbook can sound amazing with the right equipment, and if the mastering/cd production process is followed properly ie: XRCDs, listen to a 44.1 16 bit file vs a 44.1 24 bit file and you’ll definitely hear the difference.
I say no. Other than the noise floor I can't hear a difference between 8 bit and 16 bit. And at 16 bits the noise floor is so low, I can't make use of the medium's full dynamic range.
Are you saying you can't hear a difference? Like others commented above, it's quite obvious, even through fairly low-cost gear.
Where 24 bit has it over 16 bit is 120 db dynamic range vs approximately 90 db for 16 bit. This is a nice cushion when recording. Now playback is a different story, 16 bit is all you need.
So I do all my lp recording at 24/88.1 and I have to adjust levels for every recording -but going into the red (0 db and above) is still going into the red as far as I can tell. So does this extra cushion mean I can leave the clipping above 0 db? I always thought that with digital that clipping distorts the whole spectrum at that point (for whatever time it is clipping above 0 db).
I wish I could setup a 16 vs 24 bit blind test group. I think I would make a lot of money taking bets. No, there is no real audible difference between 16 and 24 bits when all else is equal. If I take a 96/24 file and resample to 44.1 and dither one to 24 and one to 16, I would be shocked if people are able to blindly tell the difference with any certainty. That said, I like to have the highest rate in which something is available as I like to remaster and feel much better about starting with 24-bit for that. Same for recording, I will definitely record in 24.
24bit/192 kHz seems to be my listening “sweet spot” but most resolutions can offer dynamite reproduction if properly recorded and mastered.
Unless using a floating bit, then you are correct that you cannot exceed 0 without clipping. The extra "cushion" is the floor being lower. But 16-bit floor is already so low that no one is hearing it at normal listening. For most of the recordings being discussed on the Hoffman board (vintage analog recordings), the noise floor on the recording is already above that of 16-bit.
No, it means you don't have to set your level so close to 0 dB. You can record at a lower level and then normalize.
So it's kind of the opposite I guess. I usually seek a minus 3,5 -4 db peak. What you describe is how I record anyway, I don't bother with normalization although I may in the future.
If you are more of a visual learner, you can also try bringing a digital photograph or a portion of a digital photograph into clipping and see how you lose spatial frequencies (the analogue of sampling across pixels instead of across time samples on an audio file).
I haven't noticed a consistent difference between a CD rip of a high quality recording, and the what I think is the same recording on Qobuz at 44.1/24. But I have no idea if the Qobuz recording was actually recorded and mastered at 24 bits, or if this is just the file format. There have been cases where I thought one was slightly better than the other, but the differences are subtle at best, and it's not always the 24bit Qobuz recording that I perceive as better.
It depends on what software you have, but adding brightness or contrast is probably the easiest way. You could also push up the various color channels if that is available. For a 24 bit color file, once either red, green, or blue are at 255 you are at the clipping threshold.
That's good to know. I used to use Photoshop all the time but I haven't touched it in years, I am a visual artist but the main focus is outdoor pictorial painting with most color matching done by eye.
Other way around. It is nearly impossible to detect. Blind A/B tests (myself and many others have taken) back that up. Archamigo needs to chime in....
You'd still lose money, as you'll need $50K worth of digital gear in DACs and transports to hear the difference.......
