Analog sound in digital world – Part 1: Noise

There are several reasons why people naturally tend to prefer sound coming out from analog gear. Getting into analog hardware is quite expensive and most people doing music out there are running on a low budget. But this isn’t something that can stop you to sound more analog. “Analog sound in digital world” is a series of tutorial I will write over time giving you different ways on how you can get that warmth into your mix. I really hope that you will have fun  and also get satisfaction from these tutorials.


This first part is divided into two sections, the first being noise coming from circuitry and the second one noise from tape. I think it is interesting to understand a bit how tape works. But, I don’t want to go too far into theory behind tape technology. So, here is a nice and short description found in the last CM issue.

So how does tape works? Coated plastic reels of tape pick up a magnetic signal from a recording head and record them, whereupon they are played back by a play head. The faster the tape playback, the better the quality (more top-end), but playing it slower could add character and means using less tape (tape can be pricey!). Because it is a physical medium, the coating gets worn off, and the tape heads get dirty, all of which reduces quality with every playback. Limitations include hiss (background noise), wow/flutter (variation in tape speed, manifested as slight pitch modulation), and crosstalk, whereby the audio recorded onto one track would also be heard very quietly on the adjacent tracks, due to the magnetic influence of the tape heads on the tape. Tape also affected frequency/phase response, softened transients, and gave distinctive-sounding tape compression/saturation, which can be employed as a useful sonic shaping device. -Computer Music Magazine, CM188, March 2013

After reading this well popularized theory you could think that I will say to use a Scream 4 device with Damage Type parameter set to Tape into each Mix Channel. Well… yes and no !! Yes, my tape faking technique involves the use of Scream 4 with his parameter set on Tape but I’ll get a bit further by adding some fake circuitry noise to my instrument. Ok, enough talk let’s begin.


We first need to have a noise source. To create that noise source connect a Thor device in the rack to a Mix Channel. Label Thor as “White Noise” and his Mix Channel “Noise Channel”. Right click, [Control/Click] for MAC user, on your Thor device and hit the [Reset Device] option.

On your Thor device set oscillator 1 [Osc 1] as a “Noise Osc” and set the noise type to “white”. Set Filter 1 as a “State Variable Filter” and use these values:

Filter Type: HP 12
Self Osc: Off
Freq: 1.80 kHz
Env: 0
Vel: 0

Also, you need to shut off these 3 envelope generator [Filter Env/Global Env/Mod Env] and disable the “MIDI” button in [Trigger] section. To make a constant source of noise go into the sequencer section and set these parameters to these values. These values will constantly trig the Amp Env when you will play your project.

Run Mode: Repeat
Rate: 1/64
Step Count: 1

The last thing we need to do to get a sustained noise sound is to set the Amp Env to  a fast attack and a long release.

Amp Env Attack: 0
Amp Env Sustain: 0dB
Amp Env Release: 1.33 sec
or above (it will only affect the duration of the noise when you stop your song)

If you think of having missed something  just compare what you have done  with the following snapshot.


If Thor is running right now you should obtain something similar to this.

Really dull isn’t it ? Ok lets make it sounds even more natural. Put a Scream 4 device as  an insert FX into your Mix Channel previously named “Noise Channel” and set your Scream 4 with the following values.


Damage Type: Tape
Damage Control: 30
P1: 90
P2: 127
Cut: On
Cut Lo: 0
Cut Mid: 0
Cut Hi: -9
Body: Off

With this small addition our noise sounds a bit warmer and less agressive.


At this point, what you need to do is to distribute that noise into every instrument’s Mix Channel present in your project . Ok, maybe you’re asking to yourself why I want to send noise in every instrument’s channel? Why not in a group channel or directly in the master bus? Because noise will be shaped accordingly to the Mix Channel settings  resulting in a different noise coloration. The sum of every noise injected in the different channels will result in a more natural colored noise effect than if you’ve just put it into your master bus. To do so, add an Audio Spider connected to the Direct Output of your Noise Channel. I recapitulate, from Thor to Spider your overall connection should look like this.


Now, I’ll show you how to introduce and blend that noise signal with one instrument. After this, repeat these steps for every instrument in your project. You don’t have to do this for group channel or auxiliary channel. Noise will propagate through groups and auxiliaries.

Create a Line Mixer 6:2. Connect Ch.1 Left Input [Mono] to one output of your Spider Noise distributor (Red line). Connect the Ch.2 input to your instrument output (Yellow line). And finally, connect the output of your Line Mixer 6:2 to your instrument’s Mix Channel input (Green line). The picture below illustrates how to proceed.


The noise quantity entering into your signal path must be subtle, almost imperceptible. The illustration below show you the ratio between noise and instrument signal. In this example I have dropped the noise level to a very low level.


Ch.1 knob(noise level) is set to 5 while Ch.2 knob(drum level) is set to 115. In this situation I barely hear the noise, but if I mute and unmute my noise channel it is there just enough. Now, repeat these operations for your other instruments.


When you are done with every instrument you will probably notice that the overall noise level have increased. We finally adjust the overall noise level at a proper one with the noise Mix Channel’s fader. Like described previously, mute and unmute your Mix Channel to see if your noise is really there and not too loud (unless you want this). The rule here is to be careful and gentle with your adjustment. You may think that if its so subtle it won’t add anything more, but the sum of these small details shall contribute to the quality of the final result.

We could let things like this but I’ll go a bit further. You might be pissed of this tutorial but… Yes, I’m a perfectionist! At this point this noise source sounds great but I want to feign my master bus being recorded on a tape.


For the last step we will create a parallel processing unit. Why using parallel processing? Parallel processing allows you to get better control  of your processed signal. Even if you set your Scream 4 (or any other fx device) to  heavy parameter (such as Damage: 127, Tape Speed: 0, Compression: 127) resulting in a highly deteriorated signal, you can still keep the mixed sound fairly natural-sounding simply by keeping the heavy processed stream low compared to the dry stream. To create our parallel Tape processor, put in the Master Insert FX section those devices and label it as described:

1x Thor | Label: Dry
1x Scream 4 | Label: Tape
1x Line Mixer 6:2 | Label: Mix
1x MClass Maximiser |Label: Limiter

The illustration below shows you how to connect your device. The yellow line is your dry signal while the green line is your dry signal inverted being fed into the Scream 4 unit. The orange line is the output signal processed by Scream 4 algorithm. Finally, the magenta line shows you the path of both signals merged back.


I want you to notice two things. The first thing is that the limiter is used with its default parameter (no input and output gain) and is there only to prevent damage to your speaker  if your master output signal accidently get too loud. If you have never done this before, you should put one MClass maximizer as a peak protection in your Master Bus in your future project. The second thing is that when using a Scream 4 set to Tape algorithm (I haven’t investigated for other algorithms yet) you will end with some phase problem cause this algorithm invert, at his output, the signal from his input. For those who don’t know the phenomenon, when two identical signals are inverted one compared with the other one, these nullify.

This is why I’m using the inputs and outputs of a Thor device as an audio splitter instead of a Spider Audio Splitter/Merger. Now, we just need to configure our matrix to split our signal between his output 1-2 and 3-4. Because output 3-4 are feeding our Scream 4 (which invert our signal) we will set our Thor’s matrix to invert the signal from his input at output 3-4 resulting in a signal back in phase with our dry signal. Also, since we are only using Thor as an audio splitter don’t forget to disable everything else. Here’s our matrix configuration.

Source: AudioIn1 | Amount: 75 | Dest 1: AudioOut1 | Amount: -75 | Dest 2: AudioOut3
Source: AudioIn2 | Amount: 75 | Dest 1: AudioOut2 | Amount: -75 | Dest 2: AudioOut4

We can now set our Scream 4 to add some warmth to our mix. To do so we’ll only use these three parameters: [Damage Amount/P1-Tape Speed/P2-Tape Compression]. I am giving some default value to use in this example as a starting point but it’s really up to you. Because I just want a bit of warmth rather than destroying my mix I will set my “Damage Amount” to a lower level. As described at the beginning of this article, tape speed affect the quality of the top-end. Because of this I will set “P1” to a higher speed to keep a relatively good top-end quality. I finally set tape compression, “P2“, to half to get a more consistent body.


Damage Type: Tape
Damage Amout: 30
P1 Tape Speed: 105
P2 Tape Compression: 64
Cut: Off
Body: Off

If you have not do this before, decrease Ch.2 Level knob to zero and hit your spacebar to hear your original mix only. While your mix is being played back increase Ch.2 Level knob to mix your tape processed signal to your taste (remember that we just want a bit a warmth, anyway it’s still up to you). Depending on your mix and your preferences you would probably set Scream 4 parameters to different values. Keep in mind that because of the signal processing method used here you can go crazy with your Scream 4 while keeping your overall mix still sounding good by adjusting the balance of your dry and processed stream.

Here is an audio example, at the beginning this is only my mix without parallel processing. Around 15 sec you will hear the Scream 4 stream being blended with our dry signal, resulting in a warmer and a louder sounding. At 30 sec, I decrease Tape Speed resulting in a different mix coloration.


Woah!! Congratulation your are finally done with this tutorial. Happy mix warming! -Simon Janvier

Tags: , , , , , , , , , , , , ,

4 Responses to Analog sound in digital world – Part 1: Noise

  • I like this a lot, so thank you!

    I believe both the Tube and Tape are phase reversed. Have set up a new song for this and a thread, check:

  • celsus says:

    good work great stuff guy

  • Gabriel Pereira says:

    Hey man, sounded amazing, thanks for that. It could be easyer if you make a video of that by the way…


    • janvier1982 says:

      Thanks for that ! Now its way more easier to do this with Reason 7 ! I think I will make an update. I know that a video could be faster than reading. But the fact is
      that I like the reading format and I can do this one part at a time. And the other thing is that I write in english better than I speak. My first language is french.
      So it is easier to explain my idea trough this format.

  • Leave a Reply to Gabriel Pereira Cancel reply

    Your email address will not be published. Required fields are marked *


    You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>