Interesting approach.
I normally have a parallel bus for drums and bass only, The rear bus technique. Find it useful for some more rock mixes.
it will depend on what you´re mixing.
I love saturation and your template is good for that, and using only reaper stuff makes it way smoother CPU wise.
Just did, and i´m really impressed with it.
Apart from being really well designed the options for saturation you have and how they are applied in the mid/side/Left/Right/Stereo are awesome !!Its worth a try..
Just did, and i´m really impressed with it.
Apart from being really well designed the options for saturation you have and how they are applied in the mid/side/Left/Right/Stereo are awesome !!Its worth a try..
Spectre is an amazing plugin. You should try out Sai'kes squashman as it offers frequency depended saturation too.
Hey guys,
did a variation on the first try. This time i'm using the beautiful Saturation Knob by Softube and Molot compressor as a parallel. Everything else is the same Reaper stock plugins. I find this version to give more depth and width to mix. You get also more definition between instruments. I'm gonna keep making several variation, so stay tuned
P.S.: Saturation Knob and Molot are free plugins!!
Would really appreciate post-fader fx so we could run Airwindows Console system more smoothly. Would be a huge leap in ITB mixing quality for many people.
Would really appreciate post-fader fx so we could run Airwindows Console system more smoothly. Would be a huge leap in ITB mixing quality for many people.
As per my post on the last page; you don't need to wait. All you need is for someone to take Chris' open source code and make a multichannel plugin.
So what does analog summing "does" technically? It subtly adds imperfections, basically. Different frequency response between left/right channels, different phase response. Also, it degrades the signal to a certain extent. And it adds subtle harmonic distortion. All this is sublty different for every channel.
So, it all comes down to: stereo image, filtering, saturation/distortion. ITB we have EQ, Waveshaping, stereo imaging plugins.
I use satson on some channels and on every buss including the master. In a blind A/B I always choose the one with Satson. It sounds slightly deeper, slightly wider... ya know all the same stuff they say about analog summing.
Man, this is really on point. I learned a lot! I strongly suggest you to try my Summing template . Also, if you have suggestions, please share with us.
I use ReaComp for pretty much all of my saturation needs nowadays.
I made a preset called "rail" that is just all time constants at 0, ratio at infinity and knee at 6. You can mess with that knee to get harder or softer action, but I found that 6 was pretty close to the tanh curve that most people use. It also has Auto make-up checked because then the threshold is pretty much like an inverted preamp gain/trim. That's not always what I want, but it's pretty convenient.
I have another preset called "transformer" that is the same thing except with the lowpass set way low. You then have to bring the threshold way down, and it gets really nasty if you go too far, but then burning transformers sound pretty horrible too.
Of course, if you use the aux input for detection, you can do all kinds of things. Want asymmetrical saturation? Mix in a little bit of DC with the original signal on the sidechain. More complex frequency interaction? EQ the sidechain. That thing about "crosstalk distortion" could be explored by mixing in some other track going into the detector. If you want something very close to parallel, use another ReaComp to saturate the sidechain signal, though you could also use feedback detection mode (or I suppose just turn up the dry slider). Try mixing in noise, buzz, or hum on the sidechain. Put an expander on it to get something like sag.
If you want multiband saturation, use ReaXComp instead.
I use ReaComp for pretty much all of my saturation needs nowadays.
I made a preset called "rail" that is just all time constants at 0, ratio at infinity and knee at 6. You can mess with that knee to get harder or softer action, but I found that 6 was pretty close to the tanh curve that most people use. It also has Auto make-up checked because then the threshold is pretty much like an inverted preamp gain/trim. That's not always what I want, but it's pretty convenient.
I have another preset called "transformer" that is the same thing except with the lowpass set way low. You then have to bring the threshold way down, and it gets really nasty if you go too far, but then burning transformers sound pretty horrible too.
Of course, if you use the aux input for detection, you can do all kinds of things. Want asymmetrical saturation? Mix in a little bit of DC with the original signal on the sidechain. More complex frequency interaction? EQ the sidechain. That thing about "crosstalk distortion" could be explored by mixing in some other track going into the detector. If you want something very close to parallel, use another ReaComp to saturate the sidechain signal, though you could also use feedback detection mode (or I suppose just turn up the dry slider). Try mixing in noise, buzz, or hum on the sidechain. Put an expander on it to get something like sag.
If you want multiband saturation, use ReaXComp instead.
whoa...there is a LOT here, i have so many questions!
Can you explain what you mean by adding DC with the original?
Can you explain what you mean by adding DC with the original?
First you need something that can generate a DC signal. They're out there. It's like the simplest JS plugin you could write if you want to make it yourself. Just a slider that puts out samples with one static value.
There are a number of ways to mix that with the original audio signal but you want to make sure that's only on the channels you're sending to ReaComp's aux detector. This essentially moves the center of that waveform one way or another so that for example positive swinging signals distort at a lower level than negative. Since we're doing this in the sidechain, we've not going to end up with a real DC offset on the audio signal* which we might have to filter out later, so it's actually more like if the top rail was that much closer to 0 than the bottom one. Theoretically asymmetrical distortion like this "generates" more even harmonics than odd which is supposedly more pleasing, but in practice it's usually very subtle and unless you're really going way off center or really distorting the hell out of it, you may not notice the difference.
*Though it does kind look and act a lot like a DC offset. If you add all of the samples together, it will not come out to 0 - it doesn't actually average to 0 but to some other DC level. But silent portions will sit at 0, so you won't have to worry quite so much about pops and clicks on cuts and mutes and things.
First you need something that can generate a DC signal. They're out there. It's like the simplest JS plugin you could write if you want to make it yourself. Just a slider that puts out samples with one static value.
There are a number of ways to mix that with the original audio signal but you want to make sure that's only on the channels you're sending to ReaComp's aux detector. This essentially moves the center of that waveform one way or another so that for example positive swinging signals distort at a lower level than negative. Since we're doing this in the sidechain, we've not going to end up with a real DC offset on the audio signal* which we might have to filter out later, so it's actually more like if the top rail was that much closer to 0 than the bottom one. Theoretically asymmetrical distortion like this "generates" more even harmonics than odd which is supposedly more pleasing, but in practice it's usually very subtle and unless you're really going way off center or really distorting the hell out of it, you may not notice the difference.
*Though it does kind look and act a lot like a DC offset. If you add all of the samples together, it will not come out to 0 - it doesn't actually average to 0 but to some other DC level. But silent portions will sit at 0, so you won't have to worry quite so much about pops and clicks on cuts and mutes and things.
...or you could cheat and get this for $29 on sale, and have envelope and LFO control to boot, as well as oversampling.
It's not something that interests me, but for anyone wanting to save some money here is Eric Valentine showing how he made his own summing box, in a fair amount of detail. Apparently straight summing doesn't cut it; you need a few cascading stages of summing:
So many unknown factors about the gain-staging. The vocals sounds louder OTB.
You can route /emulate and put plug ins onto busses all day long, and never emulate a passive mix bus with great clean fast preamps for make up gain.
it may come close - but no cigar.
Mixing through a console is another ball game, with loads of electronics in the path all affecting the end result.
...or you could cheat and get this for $29 on sale, and have envelope and LFO control to boot, as well as oversampling.
But that completely defeats the point. I understand that it might be easier or more convenient for some, but I prefer the more modular approach using simple Reaper-native (free!) plugins.
I don’t worry about oversampling for these kinds of things. Any type of oversampling is going to undo the hard limit function of a saturator. With ReaComp, the loudest sample you will ever get out of it is predictable based on the threshold and makeup gain with no question of overshoots. Oversampled saturation can’t do that unless it also applies a hard limit after the downsampling stage. In my experience, if I’m pushing it hard enough that aliasing becomes a real issue, then I’m deliberately destroying the signal anyway. If one did want “oversampling” from ReaComp (or anything else) they could just crank up the sample rate and let Reaper itself handle the resampling.
But that completely defeats the point. I understand that it might be easier or more convenient for some, but I prefer the more modular approach using simple Reaper-native (free!) plugins.
I got into Nebula when I first saw someone mix on a SSL console, they were the first to offer console programs to recreate analog mixing itb as far as I know. Although you can't drive these as you can a real mixing desk, you do get nice coloration and instrument separation, making mixing easier. I'm downloading an update for the Alex B German mastering console as I'm typing, I'd be happy to run some of your stems through it if you're interested so you can hear what it adds
You can route /emulate and put plug ins onto busses all day long, and never emulate a passive mix bus with great clean fast preamps for make up gain.
it may come close - but no cigar.
Mixing through a console is another ball game, with loads of electronics in the path all affecting the end result.
Total agree with you ChristopherT
I've spent hundreds of hours wasted' to play around how to make my final mixdown better, let me be pejorative, my final stereo mixdown always sounds like crap, collapsed like a sh*t, finally I figured out (after my tracks were mixed by my friend by a passive analog summing mixer) : You can use nice plugins, different daw sequencer software, but ITB will never sound as good as using analog summing....like CristopherT said "it may come close - but no cigar"
Thats is pure math and physics.
A sound engineer described as follows: the digital mixing is a mathematical summing, for example 1+1=2 (exact 2). When you mix in analog: 1+1=2,05 (because your tracks are summed as sinusoidal voltages) + as well as adding beneficial harmonic distortion the “pleasant harmonics”, rounds, transparency and more spatial depth. If you understand the basics of voltage mixing you quickly realize how clear it is. imho interesting, in any case a good approach....
source: https://vintagemaker.net/analogsummingmixer/
There are three important things about how to mix in analog:
1. Voltage summing mixing
In the analog world, there are no 0s and 1s, only voltages. Those complex sinusoidal voltages includes rich harmonics and other important sonic elements. Under analog summation process are producing 1st, 2nd, 3rd etc. harmonics as well as saturation/rounding peaks which makes sound so called warmer, fatter, richer. But wait, we talk about a passive mixer - Yes, the benefits of analog summing is already available by a passive mixer too, so you can learn how to hit the pleasant harmonic distortion.
How to hit the pleasant analog harmonic distortion on passive mixer? Here is the first part of the trick.
Simply boost up your DAW output Level (master fader) – and you get some “nice” distortion because you’re pushing the Passive side of Summing Mixer quite a bit over its nominal level.
2. Headroom:
ITB not have headroom, that all we know, we need loads of limiter on tracks and master, simply no playground.
Headroom is the point “the safe place” where your transients are not damaged.
Headroom is how much room has your audio signal has before it starts to get compressed and distorted.
You have around +20dB's playground, IMHO that's more than enough
It provides buffer zone for harmonics, transients or loud sounds without risking clipping, which result a more dynamic open and wide, depth 3D sound.
read more: https://vintagemaker.net/analogsummingmixer/#headroom
3. Gain Staging and hit the “sweet spot”
So it’s a little more complex, the point is that You need to “adjust” levels from each side (DAW – OUT/SUM – OUT/ MICPRE Input – GAIN) until you find the best signal ratio / settings / sound quality. Find the sweet spot in this RANGE – where the harmonic distortion starts gently and it’s about right.
read more: https://vintagemaker.net/analogsummingmixer/#sweetspot
*an addendum to the previous three:
Panning Instruments:
Find the right place of Your instruments – where they sit better in the whole mix – analog summing it provides better separation, stereo image, so take account to find where they sounds better in to the stereo field.
Finally is NOT a plug or ad here, I am an enthusiastic admirer of vintage maker gears (saved my life) because he made me custom solutions for my own studio, he make exactly what you need, you can get an many channels and options you want, and You could get one larger or more featured gear. I already own (use or have used longer) 4 gears (1 mini desktop mixer, 1 monitor controller, 1u passive mixer with gain control and a 2u dual active/passive mixer from VM, see pictures.
ps. If you want sound colour beside of DAW interface amps add an EQ for colouring.
Adding tone can massively redefine the tone of your mix, and sounds really cool..
I use it rarely, rather i use Ozone mastering software on the already recorded / summed master stereo track.
I agree that running a singnal through hardware adds saturation/distortion. Besides that it rolls of some of the highs and lows as well. I also agree headroom can be important if you use plugins that require it. I just don't think there's any magic going on that can't be done in the box.
I agree that running a singnal through hardware adds saturation/distortion. Besides that it rolls of some of the highs and lows as well. I also agree headroom can be important if you use plugins that require it. I just don't think there's any magic going on that can't be done in the box.
Don't have to believe in the magic.
I didn't believe it either.
Until I tried it. Thats is pure physics.
It's more than a simple single stereo signal trough hardware. (signal treatment by transformer or tube saturation) that use on the end of summed (after summing mixer) master signal.
Is Voltage summing, mix in analog keyboards, daw, drum machine.
Again: boost up your DAW output Level (master fader) – and you get some “nice” distortion because you’re pushing the Passive side of Summing Mixer quite a bit over its nominal level.
The magic happens. In the box. Without box you can't sum your stems, and you cant hit the analog pleasant distortion on each instruments, even more on master summed stereo signal.
Use different external amps for different sound colour, or non colouring amp, like your daw interface amp, instrument ins and remain clean transparent.
Call me cynical but I smell the biz finding something else to sell people. Is there any blind listening test data on this? There are so many ways to add console emulation etc these days I struggle to see how more (expensive) kit does something a plug can't. I tried bouncing a mix through my decent-ish Mackie 1402 mixer as a quasi summing box but wasn't convinced it made that much difference. The Klanghelm IVGI on the main bus does some interesting things which are very similar. And it's free.
I've spent hundreds of hours wasted' to play around how to make my final mixdown better, let me be pejorative, my final stereo mixdown always sounds like crap, collapsed like a sh*t, finally I figured out (after my tracks were mixed by my friend by a passive analog summing mixer) : You can use nice plugins, different daw sequencer software, but ITB will never sound as good as using analog summing....like CristopherT said "it may come close - but no cigar"
Thats is pure math and physics.
I'm sorry, Rednex, but you've just presented a lot of misinformation. If you would like to know some more, read my column on what happens to our signals in analog hardware, you can do so here: https://www.puzzlefactory.uk/TOTB.html
All the best,
Dax.
(In advance: I'm not going to enter into an argument on this topic with you.)
Linear Mode
