AnsweredQuestion About Better Digital Resolution Recording at Higher Levels

Author
razor
Max Output Level: -59.5 dBFS
  • Total Posts : 1557
  • Joined: 2004/05/10 16:53:27
  • Location: Irvine, CA
  • Status: offline
2013/09/17 18:06:51 (permalink)

Question About Better Digital Resolution Recording at Higher Levels

Hello All--
 
I recently read about how you get better resolution when recording at higher levels (or you lose it at lower levels, more accurately), and it lead me to wonder about tracks that were recorded at a good level, but not an optimal level...
 
Will adding gain to the audio file increase the resolution before mix-down if it's not recorded from the source at the optimum level, or is that resolution gone forever? My guess would be it's gone forever, but I'd like a more knowledgeable answer.
 
I'm guessing this is a good question for Bit Flipper...

Stephen Davis
 
Cakewalk by Bandlab
Windows 7 Pro 64-Bit
ADK DAW - (out of business 2018)
Intel i7 4930K CPU
Core i7 SB-E MOBO
16 GB DDR3 RAM
7 TB Storage
Layla 3G SoundCard (11.5 ms Roundtrip Latency)
UAD-2 DSP
WaveLab 8 Pro 64-bit 
Sound Forge 10 Pro
#1
BenMMusTech
Max Output Level: -49 dBFS
  • Total Posts : 2606
  • Joined: 2011/05/23 16:59:57
  • Location: Warragul, Victoria-Australia
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/17 18:19:19 (permalink)
This is something I have believed in and still do...but there are many on this forum who disagree with my opinion and you will see them here disputing it.  One thing to remember is that each piece of outboard equipment has an optimal operating level, this is when the pre-amp or the microphone really hums.  You can hear a certain silkiness when you hit this level, if you are worrying about a signal that is too hot going into your DAW (not too hot that you clip the convertors) you will never get to this optimal level.  What u need to do and this is what I do, is use the trim within the DAW's mixer to reset the signal to the optimal level for mixing once you have recorded.  There is also a theory and it's only a theory that there is even an optimal level at which to record into your convertors.  Again it will be hotly disputed here on this forum but my recordings and mixes all seem to stack up these days and I use both these theories.
 
Ben  

Benjamin Phillips-Bachelor of Creative Technology (Sound and Audio Production), (Hons) Sonic Arts, MMusTech (Master of Music Technology), M.Phil (Fine Art)
http://1331.space/
https://thedigitalartist.bandcamp.com/
http://soundcloud.com/aaudiomystiks
#2
razor
Max Output Level: -59.5 dBFS
  • Total Posts : 1557
  • Joined: 2004/05/10 16:53:27
  • Location: Irvine, CA
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/17 18:36:09 (permalink)
If it sounds better, then that's all that matters. I guess my question is more on the physics of it. I don't want to take (more) time adding gain the decent takes if it's just going to make it louder and not add to the resolution.
 
Make sense?

Stephen Davis
 
Cakewalk by Bandlab
Windows 7 Pro 64-Bit
ADK DAW - (out of business 2018)
Intel i7 4930K CPU
Core i7 SB-E MOBO
16 GB DDR3 RAM
7 TB Storage
Layla 3G SoundCard (11.5 ms Roundtrip Latency)
UAD-2 DSP
WaveLab 8 Pro 64-bit 
Sound Forge 10 Pro
#3
robert_e_bone
Moderator
  • Total Posts : 8968
  • Joined: 2007/12/26 22:09:28
  • Location: Palatine, IL
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/17 18:36:57 (permalink) ☄ Helpfulby razor 2013/09/17 21:02:04
Once something is sampled at a particular resolution, nothing on Earth will be able to sample that again and somehow make that initial resolution better.
 
I am not talking about any device's alleged sweet spot, one way or the other - I am just referring to affecting the original sample's resolution.  If you have a cassette recording, and it is grainy, turning up the volume will not make it sound any less grainy.  It will just amplify the original sound.
 
In my opinion, 
 
Bob Bone
 

Wisdom is a giant accumulation of "DOH!"
 
Sonar: Platinum (x64), X3 (x64) 
Audio Interfaces: AudioBox 1818VSL, Steinberg UR-22
Computers: 1) i7-2600 k, 32 GB RAM, Windows 8.1 Pro x64 & 2) AMD A-10 7850 32 GB RAM Windows 10 Pro x64
Soft Synths: NI Komplete 8 Ultimate, Arturia V Collection, many others
MIDI Controllers: M-Audio Axiom Pro 61, Keystation 88es
Settings: 24-Bit, Sample Rate 48k, ASIO Buffer Size 128, Total Round Trip Latency 9.7 ms  
#4
brundlefly
Max Output Level: 0 dBFS
  • Total Posts : 14250
  • Joined: 2007/09/14 14:57:59
  • Location: Manitou Spgs, Colorado
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/17 18:58:44 (permalink) ☼ Best Answerby razor 2013/09/17 21:02:27
Yes higher peak input levels technically make better use of the available resolution if the signal has a wide dynamic range; the question is at what point it matters when recording typical sources at 24 bits. Most would say if you RMS is hovering around -18dB, you're golden. But you'd have to be recording with peaks all they way down at -48dB to even be falling into 16-bit territory. So the digital sweet spot is pretty wide, and signal to noise ratio in the acoustic environment gets to be an issue long before you run out of digital resolution at 24 bits.
 
The value of higher bit depths is in maintaining good resolution at the low end of the dynamic range. Since dynamic range is the enemy of THE ENDLESS QUEST FOR MORE LOUDNESS, most modern musicians don't need to worry about it. 

SONAR Platinum x64, 2x MOTU 2408/PCIe-424  (24-bit, 48kHz)
Win10, I7-6700K @ 4.0GHz, 24GB DDR4, 2TB HDD, 32GB SSD Cache, GeForce GTX 750Ti, 2x 24" 16:10 IPS Monitors
#5
drewfx1
Max Output Level: -9.5 dBFS
  • Total Posts : 6585
  • Joined: 2008/08/04 16:19:11
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/17 19:12:20 (permalink) ☄ Helpfulby razor 2013/09/17 21:02:14
There are 3 pieces to this:
 
1. Recording/analog-to-digital conversion: This may be what you're thinking of, but there is a major caveat - once the noise floor of your analog signal hitting the converters is more than a little bit above the resolution of the converter, adding more gain does not give you any more resolution whatsoever. Note that with 24bit conversion you can have a huge range where this is the case and effectively makes this a non issue. It's much more important to avoid clipping here.
 
2. Inside the DAW: Once your signal enters Sonar (or most other DAWs), it is converted to 32bit floating point (or higher). Under floating point math the resolution is independent of signal level. 
 
3. Exiting Sonar: When leaving Sonar either via an audio interface or exporting to a 16 or 24bit file, being reasonably close to the top of the available range can be advantageous, as much from an analog noise floor perspective as digital (because if the listener cranks up the volume in the analog domain they amplify the analog noise as well).
 
But if someone is thinking of using a limiter/maximizer to try to get more resolution they will end up distorting the loudest, easiest to hear part of the signal in order to supposedly get more resolution in the part of the signal that's almost always inaudible under typical listening conditions anyway. This is obviously not a sensible trade off to make.

 In order, then, to discover the limit of deepest tones, it is necessary not only to produce very violent agitations in the air but to give these the form of simple pendular vibrations. - Hermann von Helmholtz, predicting the role of the electric bassist in 1877.
#6
Grem
Max Output Level: -19.5 dBFS
  • Total Posts : 5562
  • Joined: 2005/06/28 09:26:32
  • Location: Baton Rouge Area
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/17 19:33:08 (permalink) ☄ Helpfulby razor 2013/09/17 21:02:43
I'll agree along with Bob in that once it's in the box, that's the resolution you have. End of story.

so if you have good trks already in the box and want them to "have more resolution", I wouldn't see any purpose in this. I would want to get the highest resolution when going in the box.

Grem

Michael
 
Music PC
i7 2600K; 64gb Ram; 3 256gb SSD, System, Samples, Audio; 1TB & 2TB Project Storage; 2TB system BkUp; RME FireFace 400; Win 10 Pro 64; CWbBL 64, 
Home PC
AMD FX 6300; 8gb Ram; 256 SSD sys; 2TB audio/samples; Realtek WASAPI; Win 10 Home 64; CWbBL 64 
Surface Pro 3
Win 10  i7 8gb RAM; CWbBL 64
#7
razor
Max Output Level: -59.5 dBFS
  • Total Posts : 1557
  • Joined: 2004/05/10 16:53:27
  • Location: Irvine, CA
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/17 21:01:47 (permalink)
Yup--that's what I thought. I appreciate all the answers and they make total sense. Fortunately none of my tracks are at such a low level as to degrade their quality, it's just that I started recording this particular project prior to reading about the whole bit depth/resolution thing.
 
Now I watch my record levels closer when tracking--so I get the best quality digital has to offer.
 
I lot of helpful answers--thanks!

Stephen Davis
 
Cakewalk by Bandlab
Windows 7 Pro 64-Bit
ADK DAW - (out of business 2018)
Intel i7 4930K CPU
Core i7 SB-E MOBO
16 GB DDR3 RAM
7 TB Storage
Layla 3G SoundCard (11.5 ms Roundtrip Latency)
UAD-2 DSP
WaveLab 8 Pro 64-bit 
Sound Forge 10 Pro
#8
Grem
Max Output Level: -19.5 dBFS
  • Total Posts : 5562
  • Joined: 2005/06/28 09:26:32
  • Location: Baton Rouge Area
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/17 21:10:08 (permalink)
razor
Yup--that's what I thought. I appreciate all the answers and they make total sense. Fortunately none of my tracks are at such a low level as to degrade their quality, it's just that I started recording this particular project prior to reading about the whole bit depth/resolution thing.
 
Now I watch my record levels closer when tracking--so I get the best quality digital has to offer.
 
I lot of helpful answers--thanks!




Razor I learned long ago to get the best possible signal when recording. I got this from recording at home on tape decks of various kinds. If you didn't get a good hot signal in the beginning, the noise floor would eat you alive towards the end!
 
But that said, if your trks are just too low in volume, you can try to add some gain and see if it accomplishes what you are going for. I know I sometimes do this just to see what I will get out of it. And most times, the noise floor for that trk is low enough to allow for some gain increase.
 
Just give it a try and see.

Grem

Michael
 
Music PC
i7 2600K; 64gb Ram; 3 256gb SSD, System, Samples, Audio; 1TB & 2TB Project Storage; 2TB system BkUp; RME FireFace 400; Win 10 Pro 64; CWbBL 64, 
Home PC
AMD FX 6300; 8gb Ram; 256 SSD sys; 2TB audio/samples; Realtek WASAPI; Win 10 Home 64; CWbBL 64 
Surface Pro 3
Win 10  i7 8gb RAM; CWbBL 64
#9
Bristol_Jonesey
Max Output Level: 0 dBFS
  • Total Posts : 16775
  • Joined: 2007/10/08 15:41:17
  • Location: Bristol, UK
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/18 04:42:50 (permalink)
There is a thoroughly decent article in this months SOS written by Matt Houghton on the whole subject of Gain Staging.
 
It's well worth a read if your current knowledge is a little sketchy.

CbB, Platinum, 64 bit throughout
Custom built i7 3930, 32Gb RAM, 2 x 1Tb Internal HDD, 1 x 1TB system SSD (Win 7), 1 x 500Gb system SSD (Win 10), 2 x 1Tb External HDD's, Dual boot Win 7 & Win 10 64 Bit, Saffire Pro 26, ISA One, Adam P11A,
#10
Jeff Evans
Max Output Level: -24 dBFS
  • Total Posts : 5139
  • Joined: 2009/04/13 18:20:16
  • Location: Ballarat, Australia
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/18 04:54:27 (permalink)
It was more a 16 bit thing, does not apply now with 24 bit resolutions. It is silly to record at too high a level . You won't gain a thing other than the possibilty of clipping the signal because you are too close to 0 dB FS!
 
Choose  K system ref level and work there.

Specs i5-2500K 3.5 Ghz - 8 Gb RAM - Win 7 64 bit - ATI Radeon HD6900 Series - RME PCI HDSP9632 - Steinberg Midex 8 Midi interface - Faderport 8- Studio One V4 - iMac 2.5Ghz Core i5 - Sierra 10.12.6 - Focusrite Clarett thunderbolt interface 
 
Poor minds talk about people, average minds talk about events, great minds talk about ideas -Eleanor Roosevelt
#11
razor
Max Output Level: -59.5 dBFS
  • Total Posts : 1557
  • Joined: 2004/05/10 16:53:27
  • Location: Irvine, CA
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/18 10:55:14 (permalink)
Bristol_Jonesey
There is a thoroughly decent article in this months SOS written by Matt Houghton on the whole subject of Gain Staging.
 
It's well worth a read if your current knowledge is a little sketchy.




Couldn't find it online. Was it just in the hard copy? Any links?
 
Thanks,

Stephen Davis
 
Cakewalk by Bandlab
Windows 7 Pro 64-Bit
ADK DAW - (out of business 2018)
Intel i7 4930K CPU
Core i7 SB-E MOBO
16 GB DDR3 RAM
7 TB Storage
Layla 3G SoundCard (11.5 ms Roundtrip Latency)
UAD-2 DSP
WaveLab 8 Pro 64-bit 
Sound Forge 10 Pro
#12
stxx
Max Output Level: -82 dBFS
  • Total Posts : 406
  • Joined: 2010/01/31 17:32:02
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/18 11:17:21 (permalink)
Short answer: If its analog, levels matter, it its digital, especially at 24 bit, not so much any more.    I just read a good article on this in Sound on Sound September (US issue pg 136).  This should be mandatory reading as its very informative.  Digital level and gain staging not longer need to be at the the highest you can get.   However, gain staging is critical when summing many tracks.   Analog equipment and levels however still are seriously impacted by the level and the sound will change depending how you drive that equipment.   Plugins that emulate analog equipment also depend on the levels but basic channel levels like within Sonar per channel do not and you do not need to push the levels that high.   
#13
Jim Roseberry
Max Output Level: 0 dBFS
  • Total Posts : 9871
  • Joined: 2004/03/23 11:34:51
  • Location: Ohio
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/18 11:20:28 (permalink)
Jeff Evans
It was more a 16 bit thing, does not apply now with 24 bit resolutions. It is silly to record at too high a level . You won't gain a thing other than the possibilty of clipping the signal because you are too close to 0 dB FS!



Agreed...
Also, when summing (mixing) multiple tracks close to full code, you'll have to attenuate the level of each track.   

Best Regards,

Jim Roseberry
jim@studiocat.com
www.studiocat.com
#14
gswitz
Max Output Level: -18.5 dBFS
  • Total Posts : 5694
  • Joined: 2007/06/16 07:17:14
  • Location: Richmond Virginia USA
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/18 17:40:29 (permalink)
To add what everyone is saying, you also have to watch the gain staging into your FX. It's a bad idea to make a recording and normalize the recording so every channel uses every available bit. In this scenario, you may end up distorting as you run through FX that are modeled after real units. They also model the behavior when the signal is too hot.
 
:-)
 

StudioCat > I use Windows 10 and Sonar Platinum. I have a touch screen.
I make some videos. This one shows how to do a physical loopback on the RME UCX to get many more equalizer nodes.
#15
bitflipper
01100010 01101001 01110100 01100110 01101100 01101
  • Total Posts : 26036
  • Joined: 2006/09/17 11:23:23
  • Location: Everett, WA USA
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/18 20:11:38 (permalink)
brundlefly summed it up perfectly when he said "the digital sweet spot is pretty wide, and signal to noise ratio in the acoustic environment gets to be an issue long before you run out of digital resolution at 24 bits".
 
You can think of each bit as 6dB of dynamic resolution. Not using the topmost bit means decreasing the signal-to-noise ratio by 6dB (don't anybody nitpick over my numbers here, this is rule-of-thumb stuff). With 24-bit audio, that means you've raised your theoretical noise floor from -144dB to -138dB, which is still WAY below the threshold of audibility, and way below the analog noise you added yourself when you recorded your tracks.
 
To put that into perspective, your favorite tape-recorded records, including those that won Grammys for engineering, won't have a noise floor below -70dB. You can "lose" a lot of bits (10 or 11 of 'em easily) before you're as noisy as those Grammy-winning records!
 
One other thought...analogies to analog equipment don't work. It's true that every amplifier has a "sweet spot", but that mainly has to do with the (non)linearity of transistors and tubes, as well as staying sufficiently above the noise floor. Inside the computer, any non-linearity is an intentional effect, and the noise floor is so far down it's a non-issue.


All else is in doubt, so this is the truth I cling to. 

My Stuff
#16
SuperG
Max Output Level: -63 dBFS
  • Total Posts : 1371
  • Joined: 2012/10/19 16:09:18
  • Location: Edgewood, NM
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/18 20:46:55 (permalink)

your favorite tape-recorded records, including those that won Grammys for engineering, won't have a noise floor below -70dB

 
...and yer everyday 16 bit sample-size tops that by 26db...almost like the headroom on an analog console...
 


laudem Deo
#17
The Maillard Reaction
Max Output Level: 0 dBFS
  • Total Posts : 31918
  • Joined: 2004/07/09 20:02:20
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/19 06:30:11 (permalink)
Some one mentioned the practical reality of acoustic noise floors.
 
Some one mentioned the great benefits of 32bit floating point math.
 
Those are the two important things to appreciate.
 
 
 
I like to record with full input resolution during the A to D conversion regardless of the acoustic noise floor. Then I can make full use of a noise reduction process in 32bit (or 64bit) floating point to make it sound less noisy.
 
That way I can savor the difference between POWr1 and POWr3 dither when I export to 128kbs .mp3
 
 
best regards,
mike
 
 
 
edit grammar 
post edited by mike_mccue - 2013/09/19 20:41:51


#18
bitflipper
01100010 01101001 01110100 01100110 01101100 01101
  • Total Posts : 26036
  • Joined: 2006/09/17 11:23:23
  • Location: Everett, WA USA
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/19 13:23:59 (permalink)
 Mike, I discovered long ago that subtle humor is lost on this forum. If it weren't for the fact that I can count on you and drewfx1 to get the joke, I would never bother.
 
Anyhow, everybody knows UV-22 is vastly superior.


All else is in doubt, so this is the truth I cling to. 

My Stuff
#19
Jim Roseberry
Max Output Level: 0 dBFS
  • Total Posts : 9871
  • Joined: 2004/03/23 11:34:51
  • Location: Ohio
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/19 13:55:14 (permalink)
Upon close examination,
POWr algrorithm 1 and UV22HR produce very similar results    

Best Regards,

Jim Roseberry
jim@studiocat.com
www.studiocat.com
#20
drewfx1
Max Output Level: -9.5 dBFS
  • Total Posts : 6585
  • Joined: 2008/08/04 16:19:11
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/19 14:28:52 (permalink)
bitflipper
 Mike, I discovered long ago that subtle humor is lost on this forum. If it weren't for the fact that I can count on you and drewfx1 to get the joke, I would never bother.
 
Anyhow, everybody knows UV-22 is vastly superior.




I have my own noise shaped dither algorithm I'm working on in Reaktor*. I'm thinking of calling it PR99DoubleUltraHDx3.
 
 
 
 
 
*You might find it amusing that this part isn't a joke. 

 In order, then, to discover the limit of deepest tones, it is necessary not only to produce very violent agitations in the air but to give these the form of simple pendular vibrations. - Hermann von Helmholtz, predicting the role of the electric bassist in 1877.
#21
The Maillard Reaction
Max Output Level: 0 dBFS
  • Total Posts : 31918
  • Joined: 2004/07/09 20:02:20
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/19 20:42:17 (permalink)
:-)


#22
Chregg
Max Output Level: -51.5 dBFS
  • Total Posts : 2382
  • Joined: 2010/02/22 06:14:27
  • Location: Perth, Scotland
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/20 09:14:58 (permalink)
"One thing to remember is that each piece of outboard equipment has an optimal operating level" +1
#23
Grem
Max Output Level: -19.5 dBFS
  • Total Posts : 5562
  • Joined: 2005/06/28 09:26:32
  • Location: Baton Rouge Area
  • Status: offline
Re: Question About Better Digital Resolution Recording at Higher Levels 2013/09/21 01:55:46 (permalink)
bitflipper
I discovered long ago that subtle humor is lost on this forum.

 
I was going to say something subtle, but decided you were right.
 
[edit][/edit]
post edited by Grem - 2013/09/21 11:00:45

Grem

Michael
 
Music PC
i7 2600K; 64gb Ram; 3 256gb SSD, System, Samples, Audio; 1TB & 2TB Project Storage; 2TB system BkUp; RME FireFace 400; Win 10 Pro 64; CWbBL 64, 
Home PC
AMD FX 6300; 8gb Ram; 256 SSD sys; 2TB audio/samples; Realtek WASAPI; Win 10 Home 64; CWbBL 64 
Surface Pro 3
Win 10  i7 8gb RAM; CWbBL 64
#24
Jump to:
© 2025 APG vNext Commercial Version 5.1