Go Back   Cockos Incorporated Forums > REAPER Forums > REAPER General Discussion Forum

Reply
 
Thread Tools Display Modes
Old 05-09-2015, 10:48 AM   #1
frkasper
Human being with feelings
 
frkasper's Avatar
 
Join Date: May 2015
Location: São Paulo, Brazil
Posts: 189
Default Assessing noise levels of the current gear.

Dear all,

I'm new in this forum and a happy REAPER user and am quite excited to exchange information. I have posted the same doubt in my native language too (here).

Music/Recording is my newest hobby and I tend to get the best and leverage as much information as possible with the status-quo, i.e., the current gear I have.

I have a Focusrite 2i4 plugged into USB and have made a simple experience, given as follows:
  1. Without anything plugged in the interface inputs, and switching the knobs to no gain (0dB), I recorded a clean signal in REAPER (e.g: 5s in 44.1kHz and 24-bit PCM);
  2. Then I have normalized the item and found a 90.3dB gain (highlighted in the picture below). I read this value as a magnitude order of the noise in my system, i.e., interface + notebook and environment:
  3. If I play/send this clean signal to a FFT analyzer I can figure out what frequencies are generating the noise and under what levels. I used SPAN for that and saw that the visible -- under a -126dB floor and yet not audible -- started to kick in after 2kHz. Another picture is given below:
And why am I worried about this? You know, if I eventually need to amplify something recorded (e.g.: a low input microphone level) all these frequencies also get amplified and a hiss might be heard. I am aware that an EQ ou ReaFIR can be used here, but for the moment, I would just like to understand if a -90dB noise level is something normal or is there some issue with my gear/system.

Sadly, I don't have other gear to compare with and thought that if one can share their noise level numbers it would be good to me to understand and draw my own conclusions. Not to mention we can help each other as a community too.

The simple procedure above was the way I found to measure quantitatively this noise. Don't mean to stand this as a best practice though. On the contrary, if one have better procedures I'd be extremely happy to find out.

Please let me know your inputs and many thanks in advance.

Have fun!
Fabio
frkasper is offline   Reply With Quote
Old 05-09-2015, 12:54 PM   #2
DVDdoug
Human being with feelings
 
Join Date: Jul 2010
Location: Silicon Valley, CA
Posts: 2,779
Default

A peak around -90dB is very good.

If you adjust your playback volume to normal play back levels (even loud playback levels) with a 0dB normalized audio file, you shouldn't hear any difference between the -90dB noise an dead-silence. (There may be some audible noise in the playback electronics, do you don't actually get dead silence when you pause playback or when you play a dead-silent digital file.

Quote:
Without anything plugged in the interface inputs
An open input preamp may be noisier than a shorted input than with a low-impedance source connected to the input.

Quote:
You know, if I eventually need to amplify something recorded (e.g.: a low input microphone level)
Once you plug-in a microphone you'll find that the room /ambient noise is much worse than -90dB. That's probably true even in a "soundproof" studio. (The low impedance of the microphone might reduce the preamp noise, but acoustic noise will dominate.)
DVDdoug is offline   Reply With Quote
Old 05-12-2015, 05:14 PM   #3
frkasper
Human being with feelings
 
frkasper's Avatar
 
Join Date: May 2015
Location: São Paulo, Brazil
Posts: 189
Default

Dear DVDdoug,

Thank you very much for your valuable inputs.

I've actually tried connecting a low-impedance source into the input (I assume my portable mp3-player is one) and the noise level was quite similar than the one shown in the picture.

What I am very keen to understand is finding out why there is this build up after 2kHz towards higher frequencies (e.g.: 22kHz), seen in the FFT.

It just came in through mail my Radial ProD2 passive DI. I recorded my very hot guitar with active pickups through the XLR output and Focusrite input, and then found that I had lost 20 valuable dBs in the DI box itself (wish I knew that before ). So that is an example where that initially unseen noise while in -90dB starts to get noticeable if I need to reamp the signal.

Many thanks in advance!
Fabio
frkasper is offline   Reply With Quote
Old 05-12-2015, 05:55 PM   #4
Judders
Human being with feelings
 
Join Date: Aug 2014
Posts: 11,044
Default

The DI box is making your signal suitable for line level input. That's why it attenuates it.

I suspect the noise of a guitar amp will effectively mask any inaudible thermal noise from the input signal. Not to mention act as a handy band pass filter

In these halcyon days of 24 bit recording, you really don't have to worry about it. Just be thankful you have equipment that is so quiet!
Judders is offline   Reply With Quote
Old 05-13-2015, 01:12 AM   #5
squidward
Banned
 
Join Date: Dec 2014
Posts: 62
Default

Quote:
Originally Posted by frkasper View Post

What I am very keen to understand is finding out why there is this build up after 2kHz towards higher frequencies (e.g.: 22kHz), seen in the FFT.
span defaults to a 4.5 db/oct slope, so it tilts like yo mamma's titties !

Last edited by squidward; 06-10-2015 at 11:59 AM.
squidward is offline   Reply With Quote
Old 05-13-2015, 09:11 AM   #6
Bristol Posse
Human being with feelings
 
Bristol Posse's Avatar
 
Join Date: Jan 2011
Location: Southern California
Posts: 642
Default

Quote:
Originally Posted by frkasper View Post
Sadly, I don't have other gear to compare with and thought that if one can share their noise level numbers it would be good to me to understand and draw my own conclusions. Not to mention we can help each other as a community too.
Converter noise floor comes out at around -100 dBu
The noise floor of my 32 channel desk is around -86 dBu
My noisiest piece of recording gear is a compressor with a tube stage and the noise floor on that is around -80 dBu

however all of that is completely swamped by either room noise or guitar amp noise as soon as I turn on the mics

The ambient noise in my room is around -70dBu RMS with the mic having enough gain to get a vocal to line level
My tube guitar amps just plugged in and warmed up with no playing happening with a mic on them with enough gain to get to line level produce around -55dBu RMS

none of this is a concern to me however even on my noisiest guitar amp, when I start playing, the signal level I'm shooting for is around +4dBu line level RMS
So the signal is around 60dB above even the worst noise in my chain

You need to remember that dB is a logarithmic scale. The signal being 60dB above the noise means it has an amplitude about 1000 times greater than even the worst of the noise. You simply can't hear the noise as soon as the music kicks in. It's overwhelmed by the sounds I want to record

Unless you are getting noise that is actually audible with a real signal present without 100dB of gain....worry about something more important

Where any of this would show up in the REAPER meter would be dependent on how I calibrate the converters but it's not really relevant since the Signal to Noise ratio does not change

Last edited by Bristol Posse; 05-13-2015 at 09:20 AM.
Bristol Posse is offline   Reply With Quote
Old 05-15-2015, 11:58 AM   #7
SaulT
Human being with feelings
 
Join Date: Oct 2013
Location: Seattle, WA
Posts: 876
Default

Quote:
span defaults to a 4.5 db/oct slope, that is why you see a 'tilt' or 'slant' to the spectrum.
set the slope to '0' for an accurate reading.
Did not know, excellent!
SaulT is offline   Reply With Quote
Old 05-16-2015, 06:58 AM   #8
frkasper
Human being with feelings
 
frkasper's Avatar
 
Join Date: May 2015
Location: São Paulo, Brazil
Posts: 189
Default

Dear all,

Many many thanks again for your valuable inputs. As I tried to mention in the beginning, this is all new to me and my only reference for the time being is dBFS, given my digital gear. I'm trying to build expertise while having fun as time goes by.

Given what I read in the comments there is no more excuse. Need to record and mix something on my own very soon to eventually figure out that any audible noise will be surpassed by other sounds.

Dear squidward, that was exactly what was happening. Set that slope to 0 in SPAN.

To finalize, I thought it would be good to show a similar picture in a broader spectrum. Based on the learnings in this thread I guess I'm quite happy with the current gear. And God bless my very quiet active pickups.



Please be my guest if one wants to take this discussion further.

Have fun!
Fabio
Attached Images
File Type: jpg NothingPlugged.jpg (55.7 KB, 457 views)
frkasper is offline   Reply With Quote
Old 05-16-2015, 08:18 AM   #9
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 29,260
Default

Concerning the slope... Be aware the slope is there to compensate for how we hear which is important. It's similar to the same reason we use pink noise vs white when adjusting the response of a room in a live situation, both of which you might find interesting viewed in SPAN with various slopes. In other words, pink noise with SPAN set to a 3dB slope yields a visually flat curve that also agrees with the ears. White noise with a 0dB slope looks flat but doesn't sound flat yet is accurate for the numbers vs. the ears.

If we were to view a curve that sounded reasonably flat to our ears, we'd need a 3db slope to achieve that. Also IIRC, if you compare SPAN and ReaEQ you may notice some discrepancies in their slopes or at least which one I think should be used.
__________________
Music is what feelings sound like.

Last edited by karbomusic; 05-16-2015 at 08:26 AM.
karbomusic is online now   Reply With Quote
Old 05-17-2015, 07:52 PM   #10
SaulT
Human being with feelings
 
Join Date: Oct 2013
Location: Seattle, WA
Posts: 876
Default

Quote:
And God bless my very quiet active pickups.
Ditto! My guitars either have active pups or an onboard buffer preamp. Makes a world of difference in terms of noise and fidelity. After recording my bass DI I've now gotten into the habit of recording guitar DI as well and adding any effects after the fact. Even with hi-gain tones. It's also informative to hear when my technique and rhythm aren't up to snuff, because poor technique can just as easily be hidden by layers of distortion.
SaulT is offline   Reply With Quote
Old 05-18-2015, 09:33 AM   #11
tgraph
Human being with feelings
 
tgraph's Avatar
 
Join Date: Jun 2014
Location: Silver City, NM
Posts: 526
Default

Quote:
Originally Posted by SaulT View Post
It's also informative to hear when my technique and rhythm aren't up to snuff, because poor technique can just as easily be hidden by layers of distortion.
isn't that the point of a distortion stomp box???
tgraph is offline   Reply With Quote
Old 05-18-2015, 09:46 AM   #12
nofish
Human being with feelings
 
nofish's Avatar
 
Join Date: Oct 2007
Location: home is where the heart is
Posts: 12,096
Default

Quote:
Originally Posted by karbomusic View Post
(...)
or you just could have linked here

http://forum.cockos.com/showthread.php?t=111819
nofish is offline   Reply With Quote
Old 05-18-2015, 09:49 AM   #13
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 29,260
Default

Quote:
Originally Posted by nofish View Post
or you just could have linked here

http://forum.cockos.com/showthread.php?t=111819
I know right? I knew it was there somewhere but didn't have time to track it down so thank you for doing so.
__________________
Music is what feelings sound like.
karbomusic is online now   Reply With Quote
Old 05-18-2015, 09:54 AM   #14
nofish
Human being with feelings
 
nofish's Avatar
 
Join Date: Oct 2007
Location: home is where the heart is
Posts: 12,096
Default

Quote:
Originally Posted by karbomusic View Post
I know right? I knew it was there somewhere but didn't have time to track it down so thank you for doing so.
Not trying to mock or anything, just trying to be a little funny.
This was a helpful post for me btw. that's why I still remembered it.
nofish is offline   Reply With Quote
Old 05-19-2015, 02:19 PM   #15
frkasper
Human being with feelings
 
frkasper's Avatar
 
Join Date: May 2015
Location: São Paulo, Brazil
Posts: 189
Default

Quote:
Originally Posted by karbomusic View Post
Concerning the slope... Be aware the slope is there to compensate for how we hear which is important.
Thx karbomusic, I'm keen to understand what you meant but I guess that will take some time for to really digest it. Still looks kindda abstract to me.
frkasper is offline   Reply With Quote
Old 05-19-2015, 02:27 PM   #16
frkasper
Human being with feelings
 
frkasper's Avatar
 
Join Date: May 2015
Location: São Paulo, Brazil
Posts: 189
Default

Quote:
Originally Posted by SaulT View Post
After recording my bass DI I've now gotten into the habit of recording guitar DI as well and adding any effects after the fact.
Btw SaulT, do you use a passive or an active DI? If it's the former, have you tracked any signal loss? E.g.: my passive Radial DI loses 20dB after passing the transformer (XLR out). I am considering recording the thru signal in the interface and sending the XLR (and adapter) to amp/pedalboard.

Don't like the idea that if I eventually need to amplify my signal I bring unwanted noise too in the process. Of course I am just foreseeing a situation that might not even exist. I think I read somewhere that is better to reduce a signal than to amplify it (as long as it's not clipped of course).

Cheers,
Fabio
frkasper is offline   Reply With Quote
Old 05-19-2015, 02:29 PM   #17
frkasper
Human being with feelings
 
frkasper's Avatar
 
Join Date: May 2015
Location: São Paulo, Brazil
Posts: 189
Default

Quote:
Originally Posted by nofish View Post
This was a helpful post for me btw. that's why I still remembered it.
Looks like I have homework. Thanks for the heads up.
frkasper is offline   Reply With Quote
Old 05-19-2015, 02:37 PM   #18
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 29,260
Default

Quote:
Originally Posted by nofish View Post
Not trying to mock or anything, just trying to be a little funny.
This was a helpful post for me btw. that's why I still remembered it.
No worries, at all. Glad you found it.
__________________
Music is what feelings sound like.
karbomusic is online now   Reply With Quote
Old 05-19-2015, 06:57 PM   #19
SaulT
Human being with feelings
 
Join Date: Oct 2013
Location: Seattle, WA
Posts: 876
Default

Quote:
Btw SaulT, do you use a passive or an active DI? If it's the former, have you tracked any signal loss? E.g.: my passive Radial DI loses 20dB after passing the transformer (XLR out). I am considering recording the thru signal in the interface and sending the XLR (and adapter) to amp/pedalboard.
I don't use a DI at all. Sometimes I'll use my multifx pedal, but most of the time I just plug directly into the interface. Live I'll use DI boxes, e.g. acoustic guitar direct into the console is way better than being mic'd, but for recording I don't ever use them.


Quote:
I think I read somewhere that is better to reduce a signal than to amplify it (as long as it's not clipped of course).
All amplification raises the noise floor and gain in any form can potentially add noise to your signal. That's the idea behind proper gain staging - always being mindful of how much gain you're adding, not using too much, etc.
SaulT is offline   Reply With Quote
Old 05-20-2015, 02:28 PM   #20
frkasper
Human being with feelings
 
frkasper's Avatar
 
Join Date: May 2015
Location: São Paulo, Brazil
Posts: 189
Default

Quote:
Originally Posted by SaulT View Post
Live I'll use DI boxes, e.g. acoustic guitar direct into the console is way better than being mic'd, but for recording I don't ever use them.
Cool. Thanks for sharing!
frkasper is offline   Reply With Quote
Old 05-20-2015, 08:55 PM   #21
Fergler
Human being with feelings
 
Fergler's Avatar
 
Join Date: Jan 2014
Posts: 5,207
Default

-90db is excellent. If you record a dynamic mic even with low output like an SM7B that should still be plenty of noiseless audio to work with.

For reference, my Tascam unit gives me 85 db (same method as you used, recorded silence and normalized it) and I get completely useable signal out of all microphones I ever use, without encountering noise issues.

At that level, if noise does become an issue for some super-compressed audio or something, ReaFir and ReaGate will annihilate it for sure.
Fergler is offline   Reply With Quote
Old 05-21-2015, 09:44 AM   #22
frkasper
Human being with feelings
 
frkasper's Avatar
 
Join Date: May 2015
Location: São Paulo, Brazil
Posts: 189
Default

Quote:
Originally Posted by Fergler View Post
If you record a dynamic mic even with low output like an SM7B that should still be plenty of noiseless audio to work with.
Awesome. I do have a SM7B too and would be very curious to see a FFT screenshot of your favorite recording settings (with your room noise included). Can share mine too.

I mean, I'm still trying to figure out what would be a decent preamp knob setting. My chain is SM7B -> CloudLifter CL-1 -> Focusrite. To minimize hiss, currently I think need a noise floor of -108 (FFT-wise). If I increase it 10dB and need to amplify the item downstream, hiss is then noticeable. But working with the quieter noise floor my peaks (voice only) hit around no more than -24dBFS (need to confirm though).
It's funny shouting in a bedroom. What the neighbours would think, right?


One thing to note folks is that I have been recording and trying to learn without never putting altogether in a mix. I don't know if I'm being too picky about it. Can the hiss be overwhelmed by other frequencies in the mix? Theoretically, yes. In my stuff? Don't know.

Still haven't mastered ReaFir. The times I've tried it, I always got some ploc or "robot-like" sounds.

Thanks!
Fabio
frkasper is offline   Reply With Quote
Old 05-21-2015, 09:52 AM   #23
karbomusic
Human being with feelings
 
karbomusic's Avatar
 
Join Date: May 2009
Posts: 29,260
Default

Quote:
Can the hiss be overwhelmed by other frequencies in the mix?
Yep, it's called masking. Given any two instances of a frequency, the louder one will mask the quieter one as if it weren't there from a hearing perspective. This is actually a quite critical phenomenon in some audio data compression algorithms.
__________________
Music is what feelings sound like.
karbomusic is online now   Reply With Quote
Old 05-22-2015, 11:53 AM   #24
frkasper
Human being with feelings
 
frkasper's Avatar
 
Join Date: May 2015
Location: São Paulo, Brazil
Posts: 189
Default

Quote:
Originally Posted by karbomusic View Post
Yep, it's called masking.
Another one to the TO-READ list. Many thanks!
Fabio
frkasper is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 05:08 AM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.