Using Mimic for reference tracks (for mixing)

[crimsonwarlock]

This is just an idea I've been thinking about the last couple of days. Just throwing this out here to discuss the viability, or lack thereof.

There are several VST-plugins that you can buy, to work with reference tracks inside a DAW. The two most important functions in these plugins are matching the playback level of the reference tracks, and synchronizing parts of the reference tracks (verse, chorus, etc.) with the parts of your track. Until now, I've been simply loading reference tracks onto audio tracks and level-match the tracks with my mix using the channel faders. Pretty simple. But I didn't have a simple solution to sync the parts of each track, besides cutting up the reference tracks and move/copy the parts around, which is a serious hassle.

So then I started thinking about slice-mode in Mimic. It's pretty easy to load a reference track into a Mimic slot, and use slice-mode to quickly set up the parts and have the correct part play in sync with my mix by setting the slice-part notes in the sequencer. For example, having slice C1# play (and loop) the verse of the reference track, I just have to put a C1# note at the length of my own track's verse to have them sync up. It seems like a nice solution so far.

The only thing I'm still not clear about is how to work with several reference tracks. I often use 3 or 4 reference tracks for a mix. It would be great if I could do this by using the slots in Mimic, but I don't see (yet) how to set that up with the sequencer notes. Of course, I can load several Mimic instances for this, as I'm planning to make this all part of my mixing template, but if I could do this with just one Mimic (which could then hold up to eight reference tracks) would be even better.

Thoughts?

[PhillipOrdonez]

Having to trigger the reference via midi and it being limited to only one track per device, plus the timer you take matching the levels and you don't get any of the benefits that tools made for this offer... My advice is:

Just get adapter metric AB and use the time you'd save to mix instead ‍♂️

[crimsonwarlock]

Having to trigger the reference via midi and it being limited to only one track per device, plus the timer you take matching the levels and you don't get any of the benefits that tools made for this offer... My advice is:

Just get adapter metric AB and use the time you'd save to mix instead ‍♂️

Setting it up is not an issue, as most of it will be done once in my template, and level-matching can simply be automated, so also not a real problem. So far, the only issue I see is that you need to select the Mimic slots to have them play back and there is no way to control the slot selection like in a combinator or such. Also, my mix-prep takes some time anyway, and setting this up in addition adds minimal time to that.

Metric AB looks great, but it is not an option for me for several reasons. Besides that, I like to see what I can do with the tools at hand

[PhillipOrdonez]

Well, then suit yourself making your life more difficult than it needs to be

How do you automate the loudness matching?

Will you only use the same song references on every mix? Otherwise the set up will happen on every single track you mix. Sounds exhausting.


Extra tip: there's no need to reference every section. Focus on referencing the main part and the rest of the song should work. No need to use a breakdown for reference or an intro. Balance the rest of the song to your already balanced main part.

[crimsonwarlock]

How do you automate the loudness matching?

There's a nice RE that can do that. Takes just a few minutes for each reference

Will you only use the same song references on every mix? Otherwise the set up will happen on every single track you mix. Sounds exhausting.

I'm using specifically selected reference tracks for every project. That is basically a requirement for using references correctly. Referencing to a track that is completely different from your project is useless. Setting up specific reference tracks for a project takes only minutes, not exhausting in any way.

Extra tip: there's no need to reference every section. Focus on referencing the main part and the rest of the song should work. No need to use a breakdown for reference or an intro. Balance the rest of the song to your already balanced main part.

That depends a lot on the kind of music you make. In most EDM-related genres, that will work, as tracks tend to have not that much of dynamic variation over different parts. In my kind of music, though, there can be large differences between verses, choruses, bridges, etc. That is obviously the reason that ALL reference plugins (including Metric AB) support part-sync with your project.

My current setup uses audio channels for reference tracks. The only thing missing is part-sync. As far as I can see, setting this up with Mimic is only marginal more work then with a reference plugin. Because there, you also have to select the parts and link them to your song markers. Actually, just dropping one note in a MIDI-clip for each part of your project might even be less work, than setting this up in Metric AB or similar plugins.

[PhillipOrdonez]

But then you got to launch the transport on the right spot or else the midi doesn't get triggered.

Minutes (or a minute) to do the loudness matching? That's way too long.

So what kind of music do you make? And where can we listen to it?

[crimsonwarlock]

But then you got to launch the transport on the right spot or else the midi doesn't get triggered.

Not a problem, as I always loop sections during mixing.

Minutes (or a minute) to do the loudness matching? That's way too long.

Why? Mix-prep takes at least an hour, five minutes to setup reference tracks is negligible

So what kind of music do you make? And where can we listen to it?

Symphonic-progressive rock. Sometimes very dense arrangements. A typical project is 50 tracks or more, with 100 or more mixer channels including all the parallel processing going on.

You can't listen to it because (for now) I only distribute my music among family and friends. I have several SoundCloud accounts, but took down the few tracks I had there because those were all pretty old tracks, not representative of where I am now production-wise. I'm still contemplating if I will put new stuff online, and if so, how and where.

[PhillipOrdonez]

Why you ask? Precisely because every minute counts. If there's a tool that allows for quicker workflow, then it's worth it. Metric AB does the loudness matching in maybe 5 to 10 seconds instead of a minute. That will compound to a long ass time saved over years of mixing


One hour mixing prep sounds a bit much. See if you can improve your workflow somewhere if you can. How long do you take to mix one song?

[crimsonwarlock]

Why you ask? Precisely because every minute counts. If there's a tool that allows for quicker workflow, then it's worth it. Metric AB does the loudness matching in maybe 5 to 10 seconds instead of a minute. That will compound to a long ass time saved over years of mixing

Ah, I get it. You are a very successful producer whose days are booked solid. Every minute counts in your music production facility. But seriously, how many users here on the forum are actually in the situation where they need to shave minutes off a project

One hour mixing prep sounds a bit much. See if you can improve your workflow somewhere if you can. How long do you take to mix one song?

Most commercial producers need more than an hour prep, that's why they have assistants for that

My workflow is pretty optimized anyway. I took plenty of cues from professional engineers for that. You saying that one hour seems much, tells me you're definitely not at home in symphonic/progressive rock music

Mixing a project takes several days, in most cases spread over multiple weeks. And again, many pro engineers will tell you the same. Just to make something clear here, a lot of popular music is highly formulaic, so can be mixed rather fast by using basic recipes. Not so much with more orchestral/symphonic stuff. This is why mixing progressive rock albums can take months (if you look around the internet, you can find enough examples). When I talk here on the forum about using several parallel channels on a vocal-bus, the general reaction is "why do you need that?", which says it all. In my genre of music, it is basic practice to have many parallel processing channels on almost every sub-bus (drums, bass, guitars, vocals, orchestration, etc.), which makes mixing this involved.

But, to get back to my own topic, I found that one instance of Mimic playing a reference track takes only one percent CPU on my machine, so having three or four instances for reference tracks is not an issue.

[PhillipOrdonez]

For sure i understand the genre works different. All the more reason to improve the workflow everywhere! How much does an engineer charge to mix a song in that style?

[crimsonwarlock]

For sure i understand the genre works different. All the more reason to improve the workflow everywhere!

As I said, my workflow is pretty optimized, but gets optimized further all the time. I have a pretty comprehensive mixing template that streamlines the setup tremendously, but I also constantly tune and revise that template based on things I found or experienced. I recently made a combinator EQ that helps a lot with drum processing and is now set up as default in my template. However, in my genre, templating things only get you so far. A lot of mixing solutions and tricks are very track-specific and don't migrate in any way to another track. Experimentation in tracking, producing, and mixing, is the normal here.

How much does an engineer charge to mix a song in that style?

Well, pro engineers don't talk shop in public (= YT). But we are talking serious budgets in the commercial realm of this genre. How much did it cost to mix a Pink Floyd album? They used to rent a studio full-time for months to do that, so in today's money, at least a couple of 100K.

[PhillipOrdonez]

Since i know nothing of the genre, how is the market for it these days? The budgets are still the same as they were in the days Pink Floyd were putting out new material?

[crimsonwarlock]

Since i know nothing of the genre, how is the market for it these days? The budgets are still the same as they were in the days Pink Floyd were putting out new material?

There is still a fairly big market for these kinds of bands, but the production budgets are nothing compared to the seventies and eighties. Most of these bands are self-producing and, often, self-engineering (as far as I know). But that's basically the same for many genres, as most big studios that were booming back then, are now gone, and we all have access to pretty professional gear anyway. My current studio is several magnitudes more capable than the professional project-studios I recorded in back then.

Mind you, I'm not in it to make money. I've had my run back in the eighties and nineties, mostly playing live gigs as a keyboardist and vocalist, and doing fun gigs like synth demos at trade shows and a bit of session work here and there.

[PhillipOrdonez]

Awesome!

[selig]

Ah, I get it. You are a very successful producer whose days are booked solid. Every minute counts in your music production facility. But seriously, how many users here on the forum are actually in the situation where they need to shave minutes off a project

I was thinking about this recently, that the “pros” need time saving features that the hobbyists would never need to worry about. Then I was reminded about all the folks here who only have a few hours on the weekends (between job and family) to actually sit and make music, prompting me to wonder if those folks were ALSO just as time conscience as the pros? File under “shower thoughts”…

[Quarmat]

Symphonic-progressive rock.

I'm sorry to go so shamefully off-topic, but my brother is a prog rocker who has just finished writing an LP and since fans of the genre are rare (but usually very attentive) maybe you'd like to give it a listen. He is a Reaperian, so in more than a way you two have a background in common.

OK sorry for the intrusion, Quarmat over and out

[crimsonwarlock]

I was thinking about this recently, that the “pros” need time saving features that the hobbyists would never need to worry about. Then I was reminded about all the folks here who only have a few hours on the weekends (between job and family) to actually sit and make music, prompting me to wonder if those folks were ALSO just as time conscience as the pros? File under “shower thoughts”…

That might be a valid thought that is applicable to certain situations. Still, shaving a few minutes from a whole project?

My thoughts on this is not about the available time, like full-time versus working on weekends, but instead about if you are doing music to make music, or to make money (professionally or not). For me, making/producing music is a hobby, and although I did make money from that in the past, it has always been a hobby. Because of that, the mixing-stage is as much part of the artistic journey as everything else, at least for me. So why should I rush that. And as I said, my guess is that this goes for most users here on the forum.

And, as it is obvious that breaking into the music industry today (not MY goal) needs you to be stellar at what you do, trying to do things in as short a timespan as possible doesn't seem a great idea either. Quantity seldom tops quality for getting someone noticed

[stillifegaijin]

I just use Reference 5 - https://www.reasonstudios.com/shop/rack ... ference-5/

Free! Simple.

[crimsonwarlock]

I just use Reference 5 - https://www.reasonstudios.com/shop/rack ... ference-5/

Free! Simple.

Yep, that's exactly what I've done so far for the last few years But I'm switching to "AB 12 Audio In Switch" for that because of the option to name the reference tracks with the switches.

But I'm looking to include Mimic for reference-track playback because then I can sync up parts of the ref-tracks with parts of my project. Meaning that when I loop e.g. the chorus of my project, switching to a reference will play the chorus of the reference.

[PhillipOrdonez]

I got an idea that might work for you. Simple, too, only requires a single mimic instance, but also requires a bit of prep work, and might not work depending on how many tracks you want to use as reference per project.

Here goes:

1. Import all the reference tracks on a new project in Reason
2. Edit the sections you want from each track and delete the rest
3. line up all the clips you just cut in order on the timeline so no clips overlap and leave a small gat between each clip
4. bounce the master to a new audio track. make sure to create a memorable title for it
5. load your mix project and insert this newly created audio file onto a new mimic
6. on slice mode, the tricky part will be to set the sensitivity so that each clip will get a slice

Option 2:
using mimic but multi slot mode this time

repeat steps one and two, but bounce each clip as its own audio file (you can only do 8 this way) and import intoeach mimic slot.
Now, using several mimics on a combinator, programming the notes that will trigger each mimic, you can have a few mimics set up to react to different notes across the octaves, that way you can have a lot more sections to reference.

Do any of these help at all?

[crimsonwarlock]

I got an idea that might work for you. Simple, too, only requires a single mimic instance, but also requires a bit of prep work, and might not work depending on how many tracks you want to use as reference per project.

Here goes:

1. Import all the reference tracks on a new project in Reason
2. Edit the sections you want from each track and delete the rest
3. line up all the clips you just cut in order on the timeline so no clips overlap and leave a small gat between each clip
4. bounce the master to a new audio track. make sure to create a memorable title for it
5. load your mix project and insert this newly created audio file onto a new mimic
6. on slice mode, the tricky part will be to set the sensitivity so that each clip will get a slice

Option 2:
using mimic but multi slot mode this time

repeat steps one and two, but bounce each clip as its own audio file (you can only do 8 this way) and import intoeach mimic slot.
Now, using several mimics on a combinator, programming the notes that will trigger each mimic, you can have a few mimics set up to react to different notes across the octaves, that way you can have a lot more sections to reference.

Do any of these help at all?

Thanks for thinking this through. However, this is much more work than what I proposed. Editing the reference tracks (your point 2) is exactly what I want to prevent because of the amount of work it takes (a lot more than the few minutes we were talking about before). Also, your point 6 is not needed, as you can simply set sensitivity to zero, which gives NO slices, and manually insert the slice markers at the desired parts.

The whole idea of using Mimic, and why I started this topic, is that you don't need to do any editing of the reference tracks, and that setup is simple: drop in some slice markers in Mimic, drop the corresponding notes in the sequencer, and you're done. In addition to that, when using certain reference tracks across projects (as I do), setting the slice markers is even a one-time affair, as you can then save the Mimic patch and reload that in another project with the slices already set. Now route the Mimics through something like Reference-5 or the PongaSoft one I mentioned earlier, and you can switch between your project and the different references. Easy.

This is also where the power of templates comes in. I just have to setup the Mimics and the audio-switcher once in my mixing template. After I have prepared my reference tracks as Mimic patches, the only thing to do on a new mixing project is to load the preferred reference tracks (as Mimic patches), and drop in the notes for the reference parts. This will take a few minutes max. How is that for streamlining your workflow

[PhillipOrdonez]

Then you can just get it done using one mimic per reference track instead

[crimsonwarlock]

Then you can just get it done using one mimic per reference track instead

Yep, and I concluded that already:

I found that one instance of Mimic playing a reference track takes only one percent CPU on my machine, so having three or four instances for reference tracks is not an issue.