Enjoy the benefits of a premium membership!
Upgrade Account
-
TalkBass has been independent since 1998. Add your voice.
Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
Join free Log in
Want zero display ads or expanded classifieds tools? Compare plans.
Does vinyl sound better?
- Thread starter Nowhere Man
- Start date
There is no reference to noise floor on this page.
This article says nothing about noise floor either. It does have a reference to the signal to quantization noise ratio, which is not the same thing.
Vinyl does have a certain "vibe" to it (in terms of physical appearance, history, etc; not sound) but I also consider the non-degrading, copy-able, portable attributes of digital media to be "cool".
If I decrease the SNR, I increase the amount of noise relative to the signal. This raises the noise floor, based on the definition that the "noise floor is the measure of the signal created from the sum of all the noise sources and unwanted signals within a measurement system." Not technically the "same thing", but quantization noise is included in the noise floor, thus bit depth has a direct affect on the noise floor.
Source of quotation: Noise floor - Wikipedia, the free encyclopedia
No - the reference is to the signal to quantization noise ratio, which is not the audio signal to noise ratio. Quantization noise aka quantization error is basically just the rounding errors in the A to D; it's a relatively small factor.
[Invalid or Expired Link Removed] is a pretty good basic rundown, the relevant section quoted below (the bolding added by me). It correctly describes the noise floor as an essentially constant aspect of the analog signal being recorded, and the relationship between the noise floor and the input recording level (which is what makes it indirect) being compared based on the difference of the greater precision in the larger bit depth.
Greater precision gives you greater dynamic range which allows you greater distance from the noise floor. The noise floor still stays where it is, but it becomes less of an issue. Input recording level is still a controlling factor (as is the analog noise floor, obviously), which is why there is no direct relationship between bit depth and the noise floor.
"Noise Floor
The problem you run into is noise. Every audio system out there has some amount of inherent noise.
In other words, no recording system is perfectly quiet. The electrical components generate a low-level noise. Each piece of your system contributes to the noise party. All of this noise adds up, and its called the noise floor.
This doesnt even take into account any room noise that might get picked up by a microphone.
The noise floor essentially steals away some of your dynamic range. Lets say that all the noise added together was 18 dB. Thats 3 bits.
Since this noise occupies the bottom 3 bits of your system, the level of your audio needs to be recorded ABOVE 3 bits (or 18 dB), or it will be lost in the noise. So instead of having 96 dB of dynamic range, you realistically only have 78 dB (or 13 bits).
The gap between your recorded signal and the noise floor is getting smaller. This means that if you dont record your signal loud enough, youll end up hearing this noise in your recordings. On the flip-side, if you record your signal too loud (to stay well above the noise), youre in danger of clipping.
24-Bit to the Rescue!
Enter 24-bit recording, super-hero cape blowing in the wind.
Now, instead of giving your converter 16 measurement options, youre giving it 24. And if you kept your calculator our from earlier, then you know that 24 bits x 6 dB = 144 dB dynamic range!
An audiologist will tell you that our ears arent even capable of hearing a full 144 dB of dynamic range. However, having this much available dynamic range allows you to create greater separation between the recorded audio signal and the noise floor.
When you add in the 18 dB of noise we have in our make-believe system, and you drop the usable dynamic range down to 126 dB, you still have a TON of breathing room left.
Check out this diagram:
As you can see, the 16-bit system is still fairly close to the noise floor. The 24-bit system, however, towers above the noise floor, making it much less of an issue when recording.
In a 24-bit system, you dont need to record the levels super-hot, because youre signal is not nearly as likely to drop down into the noise floor. This leads to better sound quality, less noise, and less stress when recording."
Next you're going to tell me that my Monster Cables don't sound better(sarcasm)
Not at all. Depending on what they are compared to, they may sound worlds better. They tend to be well constructed, reasonably shielded cables. However, if they make a world of difference in your system, there was something screwed up to begin with. I just wouldn't pay the premium they charge for cables that say MONSTER on them.
If you would have just watched the portion of the video I told you to, you would have seen firsthand that reductions in bit depth increase the amount of noise within a signal. I can't seem to find my DSP textbooks right now, but if I recall correctly, the noise increase is because the approximation of the sampled signal is "jagged" as opposed to "smooth" (assuming a simple sinusoidal signal) which introduces additional frequencies that were not originally present in the signal, known as artifacts.
This is straight up BS. Tons of new records come out every month. There are plenty of companies still manufacturing high quality turntables.
Mastering is the key to a good recording.. The difference between formats being better is purely subjective. It depends on if you prefer the crackle and pop of vinyl (which i do) or the clean sound of a CD, there are too many variables to determine one as better as another.
I have no disagreement with that. However, that is a very small tangential factor relative to the discussion of the system noise floor and bit depth.
The practical thing that needs to be understood is that bit depth increases dynamic range and this allows us to record at lower levels without worrying about the noise floor.
Quantization noise is really not a part of that reasoning and is not a practical real world factor to the recording professional, who already has plenty of reasons to record at 24 bit depth. You're talking about a miniscule factor and I'm talking about a large factor which is directly controlled in every case by the things I mentioned, the analog noise floor and the input recording level.
This is what happens when you base your knowledge on hearsay on recording forums and webpages instead of understanding the mathematics and theory behind the technologies being used.
"In analog-to-digital conversion, the difference between the actual analog value and quantized digital value is called quantization error or quantization distortion. This error is either due to rounding or truncation. The error signal is sometimes considered as an additional random signal called quantization noise because of its stochastic behaviour." (http://en.wikipedia.org/wiki/Quantization_error)
Bit rate will not affect the noise from other sources. Your earlier post quoted: "This doesn’t even take into account any room noise that might get picked up by a microphone.
The noise floor essentially “steals” away some of your dynamic range. Let’s say that all the noise added together was 18 dB. That’s 3 bits."
What? The relative volume between what you want to record and background noise won't change with bit depth. Bit depth affects the SNR related to the quantization error. That's it.
That would go over a lot better if you were right about what you were saying. As it is, you just look like a guy desperately trying to save face.
Why are you posting this ? We agree on what quantization error is. Surely you aren't claiming it is a controlling factor for overall noise in the system ? You do understand that by definition quantization errors are limited in size to the difference between the possible discrete values for the given bit depth ? And that the analog noise floor is thousands of times larger than this ?
Right, so you're saying bit depth affects quantization error (only), and I'm saying bit depth does not have a direct relationship to noise floor. It kind of sounds like you're agreeing with me now.
Sure it does - if you change the input volume to take advantage of the greater dynamic range offered by the higher bit depth. Just like I said in the first place. Because input volume to the A to D, like the analog input noise floor, does (obviously) have a direct relationship to the overall noise floor of the system.
Record vinyl to CD and nobody can hear a difference between it and the original vinyl
'THE ANALOGUE SOUND RECORDED ONTO A CD IS PRESERVED 100%'
Invalid Link Removed
There's are incredible recordings of symphony orchestras recorded digitally, on CD, DVD, or download, that would never be achieved with vinyl. Even direct-to-disk vinyl. Digital just holds a lot more data.
'THE ANALOGUE SOUND RECORDED ONTO A CD IS PRESERVED 100%'
Invalid Link Removed
There's are incredible recordings of symphony orchestras recorded digitally, on CD, DVD, or download, that would never be achieved with vinyl. Even direct-to-disk vinyl. Digital just holds a lot more data.
I never said quantization error was the controlling factor in the noise of a system; quite the opposite. I simply stated that unless you hear noise in a 16 bit recording, you do not need 20 or 24 bits.
No, even with the lower bit depth, you could turn up the source, and lower the microphone gain to compensate. Actually, the bit depth doesn't offer more headroom, so I don't see how that would help; it simply reduces the noise present from sampling signals (especially those with amplitudes significantly lower than the A/D's maximum amplitude, thus increasing the useable dynamic range). The quantization (or bit depth) HAS NO EFFECT on the ratio of the desired signal to analog noise present before the A/D, nor does it allow for better recording techniques.
This works just fine. CDs are meant to reproduce exactly what they're given and they're good at it. If you record the output of a vinyl onto CD you're fine.
Issues start when you use tapes mastered for vinyl and put them directly onto CD.
That's absurd.
You are missing some very fundamental aspects of this topic which explain why 24 bit recording is the widely accepted standard among professionals.
The quantization noise you keep talking about, like I said before, is irrelevant to the reasons 24 bit recording is better.
Greater dynamic range allows you to record at lower levels , not higher. You have so much dynamic range to work with in 24 bit mode that you can just throw away about 12-20 db of it right off the top by lowering the input until you're averaging at (typically) -18 and peaking at -12, which is exactly what most professionals do - without any risk of getting within striking distance of the noise floor. This allows you to keep well out of clipping range even with aggressive plugin use.
Not coincidentally, this also happens to be why most modern 24 bit enabled A to D converters are calibrated so that 0 dbvu on a nominal +4 standard (professional) line = -18 dbfs or -20 dbfs after the converter.
Invalid Link Removed
In 16 bit mode you cannot do that; you must record hotter in order to preserve distance from the noise floor.
From the same link I posted earlier: note that the noise floor is the same for both 16 and 24 bit systems, because we are talking about the overall system noise floor, of which the digital side is such a small contributor it's not worth mentioning as a practical matter. The analog noise floor is what we are really concerned with.
Stare at that for a while and see if you start to get it.
That's absurd.
You are missing some very fundamental aspects of this topic which explain why 24 bit recording is the widely accepted standard among professionals.
The quantization noise you keep talking about, like I said before, is irrelevant to the reasons 24 bit recording is better.
Greater dynamic range allows you to record at lower levels , not higher. You have so much dynamic range to work with in 24 bit mode that you can just throw away about 12-20 db of it right off the top by lowering the input until you're averaging at (typically) -18 and peaking at -12, which is exactly what most professionals do - without any risk of getting within striking distance of the noise floor. This allows you to keep well out of clipping range even with aggressive plugin use.
Not coincidentally, this also happens to be why most modern 24 bit enabled A to D converters are calibrated so that 0 dbvu on a nominal +4 standard (professional) line = -18 dbfs or -20 dbfs after the converter.
Invalid Link Removed
In 16 bit mode you cannot do that; you must record hotter in order to preserve distance from the noise floor.
From the same link I posted earlier: note that the noise floor is the same for both 16 and 24 bit systems, because we are talking about the overall system noise floor, of which the digital side is such a small contributor it's not worth mentioning as a practical matter. The analog noise floor is what we are really concerned with.
Stare at that for a while and see if you start to get it.
The constant noise level shown IS from the quantization error. If the noise is not quantization noise, where does it come from? If it is delivered to the A/D (ie interference the mic picked up from a computer screen or noise from a fan running in the background), bit depth will not affect it (if you turn down the preamp gain to preserve headroom, the noise will decrease as well). I don't know of any significant digital noise sources after the A/D.
Either way, your point about headroom applies to recording. In that case, there is an advantage to have extra bits just to maintain headroom while staying above the noise floor (the noise floor being due to quantization error). However, since we are talking about the final product, a fully mastered audio CD, if you cannot HEAR the noise in a 16-bit version, there is no need to have a higher bit rate.
No, it isn't. Quantization noise would be so small you couldn't see it in that kind of chart.
it's the inherent noise floor in the analog signal.
Exactly my point - but input level to the A/D sure does affect it because if you record the signal louder, everything including the noise is louder.
me neither
it's more than a slight advantage. You've got a full 50 db more dynamic range. It means that, again, you can record at very conservative levels so you don't have any realistic risk of clipping other than through equipment problems or malpractice - and still keep that noise floor a full 100 db away from your average level.
In 16 bit recording on the other hand, you've got to push those levels higher, which means more compression, more riding faders, more worrying, and more overs - and you've still only got a lousy 70 db to work with (which is dangerously close to the dynamic range of a typical home system, ie you can probably hear some noise) even if you perfectly keep it all between 0 and -6 dbfs.
See the difference ? Not only do you have to do less signal manipulation to make sure you stay within a tiny 6 db swing while recording - that noise floor is still 30 db lower, even after you dither down your recording to 16 bits, because the noise floor is not going to move when you dither, it was analog noise that's already been converted.
Noise has gone from being a realistic concern (especially if your tracking levels aren't optimal) to a total a non issue.
Its 76dB down, from the 16 bit range. 76 dB is HUGE. I agree it is so small you can't hear it- that doesn't mean you can't see it when plotted on a logarithmic scale.
Bottom line here is that the only "noise floor" affected by bit depth IS the quantization error; you have yet to prove how it could possibly affect any other noise.
Then your SNR didn't change either way, so the bit increase gave no advantage.
If you take audio, be it from a record, from your recording interface, or whatever, and create a digital file with 16-bit depth, and hear no noise, what could you possibly gain from 24 bits? The entire point we're discussing is in the final product, a CD. Plus, you think you could hear noise 70dB below your audio? That's absurd.
