Hey great video! I haven't seen anybody else talk about this. One thing I suspect some of the comments aren't understanding is some mix engineers use multiple parallel buses as a creative tool to get a particular sound. It's not simple mix bus processing. Michael Brauer's system is probably the most complicated version I've seen. This technique seems about as accurate as you can get. The one thing that caught my eye is it seems like your key mix should be a version of the approved mix minus the parallel bus(s) compression. That way the key track hits the side chain without final compression, matching the original input a little closer. Thanks so much for illuminating an advanced technique!!!
Yep, you are spot on about catching that. Another person mentioned this in the comments, which I should pin. I’ll do that. It was my error, in thinking more about the video, and I should have used the file you both mentioned. Great eye, and thanks for watching.
Great video thanks for sharing. This is the first video I've ever seen where anyone figured out how to include mix bus compression. Now if I could just figure it out using my analog hardware compressor!
Hi @craigligman, well in all honesty I had some information from other people to help build the approach, but thank you for the kind words. You’d need A LOT of hardware (128 channel device would be awesome to see in real life)! 😂
Thanks for the tip. I have a client (music library) that wants a full mix but also stems that sums into a full mix and there shouldn't be any difference. I've used to do put all kinds of mastering chain on each stems, but seems like this will be much easier and sufficient.
Wouldn't it be better to use a "before mixbux" bounce as a key? Rather than a post mix bus? So it triggers the mixbus processing like it did in the mix?
Technically, yes. Good call out. I should have done that option. 👏🏻 The more I think about this, you could do either way. I’d A/B the results and see what is closer. Maybe even do some null tests.
Thanks so much, Cody. That makes me happy reading that. Let me know if you would like any specific topics covered. I’m brainstorming videos for next year.
I think this is a great way to deal with printing stems and including mix bus processing. The only (really big) caveat to me is that you limit yourself to using a compressor with an external sidechain. You're also not getting the effects of saturation. But, personally, I really think that this is absolutely good enough. Depending on the mix bus processing, you just can't expect the sum of the stems to null with the printed mix, nor do I think that clients really need that level of precision in the stem delivery.
I’d love to hear what you’re thinking when you say, “you’re also not getting the effects of saturation.” Whether clients desire or expect that level of detail I don’t really hang my hat on that. I want to give my best to them, and exceed their expectations. This topic isn’t often illuminated until months or years after the fact. 😃
@@shanegrush what I meant is that, unless you use a saturation plug-in like Saturn that also has an external sidechain, your isolated stems won't be driving the saturation the same way as the full mix. Consequently, you'll get different results.
@JulianDavidMusic - I’ll have to test that! Admittedly, I don’t always use saturation on my mix buss, but now you have me thinking. Thanks for highlighting that.
Thx for this video. Quick ? Could you not put a compressor on each subgroup, key that compressor (copy and paste from 2 bus) and record to their own print track simultaneously? Cheers
Hey, @mysterybro100 ! I wouldn’t recommend that approach. Why? The thing to keep in mind is that when you start messing with subgroup architecture, the input can change. Whether it’s delay compensation or how a DAW handles the single flow, you might not get a 1:1. You could do a null test and try, though. The goal here is always to have the same input and output signal flow, thus allowing the source material and what feeds the mix bus comp to stay the same and then the result is 1:1. I hope that helps. Thanks for the comment and for watching.
Wouldn’t the latency be different when having the compressor reaction to the entire mix as individual tracks vs the two track? Also, is the goal of providing stems so the client can use individual pieces of the mix for other purposes, or to allow for stem mastering? Because only the latter would really mandate mix bus processing be included in the stems.
Hey Mike, latency hasn’t been an issue for me in this context since the processing we’re discussing is on the final output. If you changed the plug-in inserts or bypassed one of them, yes, the latency could be altered. Automatic delay compensation is supposed to help in those situations, but I have low confidence in some of Avid’s processing. I imagine sample rate may play a factor as well. This is a good question, though. I will always do some tests. Print stems and do null tests to see if 1’s and 0’s are in order. It’s not an exact science, but the goal is to deliver the final files to archive the work and be thorough in supplying what files the client gets at the end of the project. I do have clients that use stems in creative ways, such as making their own content or, supporting any marketing efforts or using the stems for the live show playback. Technically, you could use the stems for stem mastering, but I’m not an advocate of stem mastering. As a mixer by trade, I want a mastering engineer to elevate my mix. To me, stem mastering means the mix could be better, and the mixer didn’t nail it. All in all stems are a great deliverable for labels (project archiving), clients that need the files for marketing, or even to use during live show playback. As technology evolves, stems serve as a way to preserve the project in a manner that will be useful moving forward. Thanks for the comment and for tuning in.
hey shane, thanks for your great video. i have to do a mono "stem key" bus in order for it to work. in the key input of the compressor, it seems i can only choose a mono bus. did you have the same experience?
Some plug-ins only allow a mono side chain input. So, that is a variable in what you’ve asked. In that case, I'll use the mono side chain input but it doesn’t affect my work negatively since my stereo buss processing is all LR linked. Even with a mono input, I trigger the compression for the L and R, and that's perfectly acceptable to my ears. The tricky part is if your preference would be to have LR unlinked. Here's how I handle it: my mix busses prior to my main stereo mix buss are all unlinked. Small caveat, I have my version of “Brauerizing” where I use multiple mix busses feeding into a main, and final mix buss.
Don’t you get better control if you just copying the mixbus settings, and bounced the stems without the mixbus processing active, and use the mixbus settings on the stem-mix? I do see the benefit due to analog gears and settings (picture), but what benefits do your workflow achieve for a total in-the-box mix?
Hey @oh515 ! I don’t think so, personally. If you’re copying inserts across stems you’re only processing that stem. You essentially lose the summed nature of the signal. Does that make sense? Am I understanding you right?
@@shanegrush Yes, it makes sense if you for a reason need to print the stems with summing, but don't you get better control later if you bounce the stems with a clean mixbus? It will be the same result if you import the stems in a new project and process the stems through the saved mixbus processing (from the main/pre mix). Or is this a technique for use in Atmos mode?
When I say “summed,” I don’t mean summing. I mean, multiple inputs are going into a single stereo aux (your mix bus) and then printed. If you have stems and identical processing on each stem, you do not get the interaction. That’s where my process comes into play. Why make an extra step for yourself? Finish the mix, and print.
I feel like this is staying true to the letter of what the suits are asking, whilst not being what the suits actually want. What they want is the option to do versions of the mix, vox up vs vox down, with drums, without drums, etc. If the vocals are pumping with the drums, but then you remove the drums, then the vocals are still pumping. This is very unlikely to be what the record company actually wants from you when they say they want stems. It would be less work for you, and more reproducible for the engineer that follows you, if you print stems without any mix bus dynamics processing enabled. The mastering engineer is going to add compression and limiting anyway, so if they want to remix from stems, the loss of your compressor settings wouldn't be a problem. Anyway, I'm not a pro, so feel free to ignore all of this.
Hello, I understand what you’re saying, but what you have to remember is that the mix bus processing isn’t pumping enough to be noticeable should one of these stems be soloed. The consumer rarely gets these stems unless there’s some remix competition or public release of them (which is becoming a thing these days). The ultimate goal is to get the mix stems to MATCH the main mix. Before this technique, it has not been possible. I’m not saying that I’ve conquered Rome here. I’m just sharing a technique I’ve created where not one business person has questioned the integrity of the work. That’s the value I hope others hear. All the best, and thanks for watching! 🤜🏻🤛🏻
@@shanegrush in which case, that's the ultimate test: if your clients are happy, then you've no need to worry about some guy commenting on the internet! I've often thought about how best to go about enabling stem mastering, luckily I just do this for fun so I've never had to. Thanks for replying, take care!
I've looked into the Internal Renderer, and I wanted to follow up on your comment. The routing is highlighted on a few videos Avid has released. Check those out! Also, the manual outlines the changes to the I/O page to help you understand the signal path when using the internal renderer. I hope this helps. Starts on Pg. 57, resources.avid.com/SupportFiles/PT/Pro_Tools_Reference_Guide_2023.12.pdf
Sorry I'm dumb, but what's the point of the comp then? It was keyed to compress an already approved track when you print the stems, not compressing the mix. What's the point of using the comp at all? Why don't just print the stems as your final mix has been approved? Your idea is good, but I think I will improvise a little to make sure the comp compresses the mix, not the approved track. But even in this case, if the final track was approved, why do I need to use a comp to alter the sound again?
You’re not dumb. The stems have to be printed at some point. I’m just highlighting how to print them with buss processing. I am showing an example that is already completed, so maybe I should have walked through an incomplete example. The stems are a separate deliverable alongside the stereo mix, and this approach allows an engineer to make a final mix, and then print very accurate stems.
Hey great video! I haven't seen anybody else talk about this. One thing I suspect some of the comments aren't understanding is some mix engineers use multiple parallel buses as a creative tool to get a particular sound. It's not simple mix bus processing. Michael Brauer's system is probably the most complicated version I've seen.
This technique seems about as accurate as you can get. The one thing that caught my eye is it seems like your key mix should be a version of the approved mix minus the parallel bus(s) compression. That way the key track hits the side chain without final compression, matching the original input a little closer.
Thanks so much for illuminating an advanced technique!!!
Yep, you are spot on about catching that. Another person mentioned this in the comments, which I should pin. I’ll do that.
It was my error, in thinking more about the video, and I should have used the file you both mentioned. Great eye, and thanks for watching.
Great video thanks for sharing. This is the first video I've ever seen where anyone figured out how to include mix bus compression. Now if I could just figure it out using my analog hardware compressor!
Hi @craigligman, well in all honesty I had some information from other people to help build the approach, but thank you for the kind words.
You’d need A LOT of hardware (128 channel device would be awesome to see in real life)! 😂
Thanks for the tip. I have a client (music library) that wants a full mix but also stems that sums into a full mix and there shouldn't be any difference. I've used to do put all kinds of mastering chain on each stems, but seems like this will be much easier and sufficient.
Yes! I’ve had those same “working” issues. This is the best way I’ve found to remedy the topic. Glad I could help, and thanks for watching.
That Drum Sound is Killer
Thanks so much @schmuckeprinz .
Clever thinking with the sidechain!
Wouldn't it be better to use a "before mixbux" bounce as a key? Rather than a post mix bus? So it triggers the mixbus processing like it did in the mix?
Technically, yes. Good call out. I should have done that option. 👏🏻
The more I think about this, you could do either way. I’d A/B the results and see what is closer. Maybe even do some null tests.
@@shanegrush Awesome. Your Dolby Atmos videos been helping me out. Good work.
Thanks so much, Cody. That makes me happy reading that.
Let me know if you would like any specific topics covered. I’m brainstorming videos for next year.
Studio one covers all of these topics
Thanks 💯💯💯💯💯💯
You’re very welcome. Thanks for tuning in.
I think this is a great way to deal with printing stems and including mix bus processing. The only (really big) caveat to me is that you limit yourself to using a compressor with an external sidechain. You're also not getting the effects of saturation. But, personally, I really think that this is absolutely good enough. Depending on the mix bus processing, you just can't expect the sum of the stems to null with the printed mix, nor do I think that clients really need that level of precision in the stem delivery.
I’d love to hear what you’re thinking when you say, “you’re also not getting the effects of saturation.”
Whether clients desire or expect that level of detail I don’t really hang my hat on that. I want to give my best to them, and exceed their expectations. This topic isn’t often illuminated until months or years after the fact. 😃
@@shanegrush what I meant is that, unless you use a saturation plug-in like Saturn that also has an external sidechain, your isolated stems won't be driving the saturation the same way as the full mix. Consequently, you'll get different results.
@JulianDavidMusic - I’ll have to test that! Admittedly, I don’t always use saturation on my mix buss, but now you have me thinking. Thanks for highlighting that.
Thx for this video. Quick ? Could you not put a compressor on each subgroup, key that compressor (copy and paste from 2 bus) and record to their own print track simultaneously? Cheers
Hey, @mysterybro100 ! I wouldn’t recommend that approach. Why? The thing to keep in mind is that when you start messing with subgroup architecture, the input can change. Whether it’s delay compensation or how a DAW handles the single flow, you might not get a 1:1. You could do a null test and try, though.
The goal here is always to have the same input and output signal flow, thus allowing the source material and what feeds the mix bus comp to stay the same and then the result is 1:1. I hope that helps. Thanks for the comment and for watching.
Wouldn’t the latency be different when having the compressor reaction to the entire mix as individual tracks vs the two track? Also, is the goal of providing stems so the client can use individual pieces of the mix for other purposes, or to allow for stem mastering? Because only the latter would really mandate mix bus processing be included in the stems.
Hey Mike, latency hasn’t been an issue for me in this context since the processing we’re discussing is on the final output. If you changed the plug-in inserts or bypassed one of them, yes, the latency could be altered.
Automatic delay compensation is supposed to help in those situations, but I have low confidence in some of Avid’s processing. I imagine sample rate may play a factor as well. This is a good question, though. I will always do some tests. Print stems and do null tests to see if 1’s and 0’s are in order.
It’s not an exact science, but the goal is to deliver the final files to archive the work and be thorough in supplying what files the client gets at the end of the project. I do have clients that use stems in creative ways, such as making their own content or, supporting any marketing efforts or using the stems for the live show playback.
Technically, you could use the stems for stem mastering, but I’m not an advocate of stem mastering. As a mixer by trade, I want a mastering engineer to elevate my mix. To me, stem mastering means the mix could be better, and the mixer didn’t nail it.
All in all stems are a great deliverable for labels (project archiving), clients that need the files for marketing, or even to use during live show playback. As technology evolves, stems serve as a way to preserve the project in a manner that will be useful moving forward.
Thanks for the comment and for tuning in.
hey shane, thanks for your great video. i have to do a mono "stem key" bus in order for it to work. in the key input of the compressor, it seems i can only choose a mono bus. did you have the same experience?
Some plug-ins only allow a mono side chain input. So, that is a variable in what you’ve asked.
In that case, I'll use the mono side chain input but it doesn’t affect my work negatively since my stereo buss processing is all LR linked. Even with a mono input, I trigger the compression for the L and R, and that's perfectly acceptable to my ears.
The tricky part is if your preference would be to have LR unlinked.
Here's how I handle it: my mix busses prior to my main stereo mix buss are all unlinked. Small caveat, I have my version of “Brauerizing” where I use multiple mix busses feeding into a main, and final mix buss.
@@shanegrush that makes sense. thank you
You’re welcome. 🤜🏻🤛🏻
Don’t you get better control if you just copying the mixbus settings, and bounced the stems without the mixbus processing active, and use the mixbus settings on the stem-mix?
I do see the benefit due to analog gears and settings (picture), but what benefits do your workflow achieve for a total in-the-box mix?
Hey @oh515 ! I don’t think so, personally. If you’re copying inserts across stems you’re only processing that stem. You essentially lose the summed nature of the signal.
Does that make sense? Am I understanding you right?
@@shanegrush
Yes, it makes sense if you for a reason need to print the stems with summing, but don't you get better control later if you bounce the stems with a clean mixbus? It will be the same result if you import the stems in a new project and process the stems through the saved mixbus processing (from the main/pre mix). Or is this a technique for use in Atmos mode?
When I say “summed,” I don’t mean summing. I mean, multiple inputs are going into a single stereo aux (your mix bus) and then printed. If you have stems and identical processing on each stem, you do not get the interaction. That’s where my process comes into play. Why make an extra step for yourself? Finish the mix, and print.
@@shanegrush
Yes, I assume I'm not there yet. I should probably start to take that step to speed up my workflow.
@@shanegrush
Thanks btw!
I feel like this is staying true to the letter of what the suits are asking, whilst not being what the suits actually want.
What they want is the option to do versions of the mix, vox up vs vox down, with drums, without drums, etc.
If the vocals are pumping with the drums, but then you remove the drums, then the vocals are still pumping. This is very unlikely to be what the record company actually wants from you when they say they want stems.
It would be less work for you, and more reproducible for the engineer that follows you, if you print stems without any mix bus dynamics processing enabled.
The mastering engineer is going to add compression and limiting anyway, so if they want to remix from stems, the loss of your compressor settings wouldn't be a problem.
Anyway, I'm not a pro, so feel free to ignore all of this.
Hello, I understand what you’re saying, but what you have to remember is that the mix bus processing isn’t pumping enough to be noticeable should one of these stems be soloed. The consumer rarely gets these stems unless there’s some remix competition or public release of them (which is becoming a thing these days).
The ultimate goal is to get the mix stems to MATCH the main mix. Before this technique, it has not been possible. I’m not saying that I’ve conquered Rome here. I’m just sharing a technique I’ve created where not one business person has questioned the integrity of the work. That’s the value I hope others hear. All the best, and thanks for watching!
🤜🏻🤛🏻
@@shanegrush in which case, that's the ultimate test: if your clients are happy, then you've no need to worry about some guy commenting on the internet!
I've often thought about how best to go about enabling stem mastering, luckily I just do this for fun so I've never had to.
Thanks for replying, take care!
Internal Pro Tools Dolby Atmos Renderer I/O Setup make the video plz
I've looked into the Internal Renderer, and I wanted to follow up on your comment. The routing is highlighted on a few videos Avid has released. Check those out! Also, the manual outlines the changes to the I/O page to help you understand the signal path when using the internal renderer. I hope this helps. Starts on Pg. 57, resources.avid.com/SupportFiles/PT/Pro_Tools_Reference_Guide_2023.12.pdf
@@shanegrush Thanks
Sorry I'm dumb, but what's the point of the comp then? It was keyed to compress an already approved track when you print the stems, not compressing the mix. What's the point of using the comp at all? Why don't just print the stems as your final mix has been approved? Your idea is good, but I think I will improvise a little to make sure the comp compresses the mix, not the approved track. But even in this case, if the final track was approved, why do I need to use a comp to alter the sound again?
You’re not dumb. The stems have to be printed at some point. I’m just highlighting how to print them with buss processing. I am showing an example that is already completed, so maybe I should have walked through an incomplete example. The stems are a separate deliverable alongside the stereo mix, and this approach allows an engineer to make a final mix, and then print very accurate stems.
@@shanegrush Thanks for the explanation. I'll try this technique in my future projects. It's cool. Thanks!
You’re welcome! Go get em.
Studio one is just a click for all of these techniques
Cool!