"3D-sounding music" mixes.

[Mark Richards]

mj, this *is* the Recording 101 forum (), and yes, please, some details. There's always something new to be learned.

Maybe start with your vision for the mix, and how you planned it out?

 

[-mjk-]

lol, man I need to look more carefully, haha. I thought this was the Mixes for Critique forum. As usual Mark, thank you for the course correction.

Geesh! lol

 

[-mjk-]

Re: Renew:

I didn't have any vision at first. Normally I do hear potential in a tune and I think "I can make that tune sound better" but when I get the tracks I have to listen to what they are saying. This is very important and something that engineers must learn in order to be useful to their clients.

I have several must-do operations before I begin any mix (I need to do a video on that). Using RX 11 Advanced, I did the following:

1. Batch converted all files to 24b 48k. My console is 24/48 and Reaper will do the conversion upon import but I like to use RX because I can also add a prefix "RX" to the processed files and I can hear them separately outside of the DAW if anything sounds off. I have both the original and process files together in my "Import" folder for that tune. I think I probably trust the precision of RX better than Reaper for that task but the truth is probably that Reaper is just as good.

2. If the files are properly named (Lead Vocal, Guitar, Bass, etc.) I will then use the Leveler Module in RX and work them in a batch or individually as needed. Nick's files are always well named so I was able to apply the correct leveling to the individual tracks before I put them into the project. RX's Leveling Module is some kind of magic and it's not the same as a compressor. It levels out the performances in a very positive way but somehow preserves the dynamics. I can also set a target level that the DAW and my console like: -18 dB. Leveled tracks sound fuller, punchier and are much easier to manage. They also require less aggressive compression, which is always a good thing.

3. I have a typical track layout on the console that I like to work with, so I add the tracks to the project and make sure that any special routing gets done and the audio comes up on the console channels how I want them. I always mix with DCAs and I set those up near the end of the setup process, when I'm ready to start managing levels.

4. I always listen to everything flat first. I do sort of a "yardstick" mix and see where things fall together. At that point I start to make adjustments.

5. I really listen not only to the technical aspects of the mix but also to the artistic aspects and I try to interpret what the artist was trying to get across with this tune. That helps a great deal because that's when I start to get the vision of the whole thing.

6. I start to build the soundstage at this point. Everything is up in the mix and now they need placement. I work with no less than 2 reverbs and often 3 or maybe more but not all applied to everything in the mix. I define the size of the room for the band and I create that with a reverb, typically a vintage type reverb. I start out with a 1 second reverb time and work with the room size, reflections, decay and pre-delay to make a credible room. Normally I use the console's FX for this, but I have a pretty good arsenal of tools for reverbs and delays in the from of plugins. I have many spatial 3D type reverbs but I don't use them as much as I thought I would when I invested in them, lol.

I have my color e-ink tablet in the studio and I have all of my studio forms on the tablet, so I can take paperless notes with full search capability. So I do all of my track sheets, mixdown and mastering sheets on the tablet with my handwriting on the PDF forms. Very easy and a fantastic way to work in the 21st century.

To be continued:

 

[Mark Richards]

Nice write-up, mj.

When I'm establishing the rough mix, like you, I'm using the "yardstick" method. [aside: if the tracking engineer has done the job right initially, the yardstick method at unity gain should get the mix into the ballpark.]

Once I have the rough mix, I let it sit for a few days so my ears are fresh when I come back to the song. I've found over the years that the first refined mix I create is usually the keeper (unless there's something glaringly out of place). I'm not sure why, but it always seems any mix I do after the first is never quite as good. So that's become my basic SOP: rough mix; step away; refine the mix, and on to mastering.

 

[Slugworth]

I decided to remove most of the Renew (80-bpm) percussion based on Laura’s suggestion.

BandLab - Make Music Online

The cloud platform where musicians and fans create music, collaborate, and engage with each other across the globe

www.bandlab.com

 

[-mjk-]

Part 2: Let the edits begin!

After the tracks were leveled and placed into the project, the next thing that I did is perhaps the most time-consuming and tedious task - which explains why nearly no one does it: I tempo mapped the entire song.

Since I use Reaper with it's unlimited capability, I have an action that when run, places a tempo marker 4 beat from the cursor. So I can put tempo markers at every measure which makes it very easy to map. From there, I determined the BPM of the song as close as I can get, and then I dragged the markers so the grid matched the tempo variations of the song. I use the kick for the standard and I made everything else line up with that.

Tempo mapping is necessary for success. If the tune sounds sloppy and not together you cannot have a good sounding record. It's worth doing if your song is worth releasing.

Next thing was to bring all the instruments together to the tempo mapped grid. I have made several modifications to Reaper to make editing life easier. First of all, I have a mode where when I click on an audio item, it creates a split at the cursor location. So I can put a split right before and right after a hit, with 2 clicks. I also modified Reaper so that if my cursor is at the top of the item, it moves the item, but if the cursor is at the bottom of the item, it slip edits the audio instead of moving the item. That makes it extremely easy to move a snare hit to the grid, for example. I cut the audio before and after a hit and slide it into place.

So, after accurately tempo mapping the song I then went though each instrument and slip edited the hits to match the grid. Remember I am not quantizing the audio, so the feel stays the same as how it was played, but it gets very much tighter as everything follows the drummer. While this takes time and can end up being a bit expensive, consider how much tighter my delivered mix/master is compared to the others where the timing was much looser.

I was during this editing process that I discovered the percussion tracks were misaligned. If I may go off on a bit of a rant here, I would like to say that this aspect was a true "separate the men from the boys" moment. Why would the percussionist continue playing in the stop sections, while then coming back in several bars late? It just didn't make sense, yet all the other mixes exhibited this issue. The only answer could be that the tracks weren't lined up. I moved the audio items so the percussion started with the other instruments, and wouldn't you know it - the stop times lined up too.

Assuming that all tracks line up and just going along with whatever happens regardless of the impact on the tune is an inexcusable rookie workflow.

In Part 3 I will outline what I did to clean up the microphone bleed in the drum tracks.

 

[Arjan P]

Interesting read! I did also listen to the MJ mix and the first thing I notice is that it indeed sounds nice and tight. @-mjk- You are absolutely right about rhythmic alignment, but I consider this the worry of the recording engineer. And I do realize that it's not easy with tracks coming from all over the globe. Maybe I missed it earlier in the coversation, but how is that done, @Slugworth ?

On a more technical level: Wow, this track is LOUD! Integrated loudness at -9.3 dB with a Loudness Range of 3.9 dB... That's EDM territory, phew! What will happen on most streaming services is that it will simply be turned down automatically (AFAIK, YouTube for instance has a reference of -12 dB). Personally, I aim for -11.5 to -12 dB, but it's a choice. BTW: True peaks are at -0.2 dB, that is a level I used to use a long time ago, but for conversions these days most mastering engineers I know (mostly from the Wavelab forums) use a safe -1.0 dB.

 

[Slugworth]

Part 2: Let the edits begin!

After the tracks were leveled and placed into the project, the next thing that I did is perhaps the most time-consuming and tedious task - which explains why nearly no one does it: I tempo mapped the entire song.

Since I use Reaper with it's unlimited capability, I have an action that when run, places a tempo marker 4 beat from the cursor. So I can put tempo markers at every measure which makes it very easy to map. From there, I determined the BPM of the song as close as I can get, and then I dragged the markers so the grid matched the tempo variations of the song. I use the kick for the standard and I made everything else line up with that.

Tempo mapping is necessary for success. If the tune sounds sloppy and not together you cannot have a good sounding record. It's worth doing if your song is worth releasing.

From just a listener’s experience and not knowing the technical aspect behind it. It’s certainly a noticeable improvement and sounds more professional.

Next thing was to bring all the instruments together to the tempo mapped grid. I have made several modifications to Reaper to make editing life easier. First of all, I have a mode where when I click on an audio item, it creates a split at the cursor location. So I can put a split right before and right after a hit, with 2 clicks. I also modified Reaper so that if my cursor is at the top of the item, it moves the item, but if the cursor is at the bottom of the item, it slip edits the audio instead of moving the item. That makes it extremely easy to move a snare hit to the grid, for example. I cut the audio before and after a hit and slide it into place.

So, after accurately tempo mapping the song I then went though each instrument and slip edited the hits to match the grid. Remember I am not quantizing the audio, so the feel stays the same as how it was played, but it gets very much tighter as everything follows the drummer. While this takes time and can end up being a bit expensive, consider how much tighter my delivered mix/master is compared to the others where the timing was much looser.

I was during this editing process that I discovered the percussion tracks were misaligned. If I may go off on a bit of a rant here, I would like to say that this aspect was a true "separate the men from the boys" moment. Why would the percussionist continue playing in the stop sections, while then coming back in several bars late? It just didn't make sense, yet all the other mixes exhibited this issue. The only answer could be that the tracks weren't lined up. I moved the audio items so the percussion started with the other instruments, and wouldn't you know it - the stop times lined up too.

Assuming that all tracks line up and just going along with whatever happens regardless of the impact on the tune is an inexcusable rookie workflow.

Concerning the percussion track I can say that the decision that Ivan made for his arrangement on this song is not his usual MO. I didn’t discuss it with him but my guess is he may have been trying something new or possibly influenced by artists that he is currently listening to. Either way I agree it sounds odd and I’m glad that MJ salvaged it.

In Part 3 I will outline what I did to clean up the microphone bleed in the drum tracks.

Looking forward to part 3.

 

[Slugworth]

Arjan,
Most of my songs originate on bass guitar. After I record to a click track or drum loop I send them to an Airgigs engineer to be quantized. I then send the quantized track or tracks if guitar is also quantized as a mix to Ivan (drummer).

Once Ivan has his drums mixed I usually decide from there which instruments (piano, horns, violins) to add. I always send my reference vocals and lyrics sheet to the singers.

From there each person sends their track to me, which I send to the engineer. The collective stems don’t always slide exactly in place.