New ReMix Stems

[Garpocalypse]

You've got the flute pretty good, the backing vocals are good -- acoustic guitar panning is good. The mellotron strings might be a bit dry and upfront, mainly dry being the issue. I used predelay and reverb to send mine back a little space-wise.

Really? Strings were hard panned 100% and reverbed with 150% width. Even though the stem was in stereo they still sounded like they were coming up the center so I offset them by a nudge. Maybe i needed a bigger nudge. Panning went something like Backup Vocals 50/50. Guitar 75/75. Strings 100/100

I love how the cowbell cuts through in this haha!! Very upfront. This is the first mix of the song I've heard where the cowbell is really easy to hear, including my own which obscures it.

That's a cowbell?

The backing vocals seem to be at a good level but the main vocal line by comparison is a bit soft volume-wise. How did you compress the bass? Needs to be compressed really hard to get a solid sound for it, so it seems a bit too dynamic but at least it's not overpowering the song, right?

I'll take another look at that. I think I was hitting around -5 or -6 GR. I used a bass plugin I had as a post processor on that one to shape the tone a little different.

How hard did you compress the piano? During the intro it's really easy to tell that it is squeezed tight. Some of that saturation you put on causes the occasional "watery" sound, I think. I noticed it especially in the cymbals. Could be any combination of the saturation / compression -- by the way, how did you compress the master? There's a weird sound around 3:38 where it goes silent briefly before the ending, almost sounds like stuff is being pushed there. Almost like you have a gate on it or something.

I saturated the strings but other than splitting the piano to mono tracks I didn't do anything else to it. For the intro I telephoned the manny and the piano. Intentionally half killing them so that the first section could sound bigger.

I did a 1/4 note cut at 3:38 and let the tail of the flute delay fill it. No gates. I most likely did some stupid stuff.

I mastered for the first time without a compressor and instead used this Boost 11 plugin that came with Sonar. First effort in trying to preserve peaks that I think I was killing during the mastering stage.

I set up compression / pre-delay / room reverb sends before starting anything in a session. That makes it really easy to apply these same effects as sends on all of the tracks. Is this something you do, or need some clarification on how to do it? And parallel compressing the drums to get that extra little bit of snap out of them. The samples as they are sound kind of dull, so there is some dolling them up involved to make them pop.

My fault for kinda going with what I had once I got into it. I didn't compare it to the original until i finished, that's when I noticed there was alot of energy from the drums that was lacking. I'll parallel compress them next time.

I could use some clarification on the sends you use. I had one reverb set up for some tracks to be sent to. Everything else was processed as a bus or individually.

And indeed, it's NOT live drums! I think it's Studio Drummer for this one, I'd have to double check sometime to make sure but it's either that or Abbey Road Modern Drummer, they're both pretty much the same. The samples try to function like a real drum setup, that's why it has the room mics and all that kind of stuff.

Crap, and here I thought I just got some real kit experience. The room track was a reamp then?

Thanks for the chance to mix this!

Edited September 2, 2013 by Garpocalypse

[Magnetic Ether]

I'm late to the party, but I wanted to give this a shot so I spent a few hours on it this afternoon.

https://app.box.com/s/qw5e9p2fabpaoc8nhl71

Some things worked out better than others I think. It was my first time mixing vocals into a finished track so that was interesting. It took a while to work with all the backing vocals, but I think I was able to fit them in decently in the end.

I'm not too happy with the drums. Those cymbals were especially difficult to work with.

I hadn't listened to the original mix for a couple of weeks before I started and I hadn't listened to any of the versions posted here (so it's kind of funny that I did basically the same thing Gar did to the intro). Listening to them now, the biggest thing that stands out for me is how much brighter the original is in comparison to mine.

Definitely feel similar to Gar in regards to loss of energy in my mix. Especially during the sax solo, I don't know how you were able to make those guitar chords so full and present...

Anyway, thanks for this opportunity Brandon! Was interesting to tackle someone else's work for once instead of mixing and arranging at the same time.

[Garpocalypse]

I'm not too happy with the drums. Those cymbals were especially difficult to work with.

If I remember correctly the stems didn't have a cymbal track and were located only on the overheads. This might be the way most live drums are i honestly have no idea but it did make the cymbals tougher to work with.

Definitely feel similar to Gar in regards to loss of energy in my mix. Especially during the sax solo, I don't know how you were able to make those guitar chords so full and present...

The drums in the original were really compressed. Much more than I guessed the first time around.

As for the guitar chords I don't know if you were talking about mine or brandon's (probably brandon's) but I can tell you what I did with them.

-Hard pan the guitars and send them to a stereo bus with it's own EQ.

-I can't remember the filter settings exactly but I think I HPF'ed around 160-180 with a Q of about 1.0. Also a slight dip around 500hz (more out of habit). Then around 2k-6k I used a shelf to make a wide dip around the vocals. Not a hard dip, only about -5db or a little less but it's a wide one. Then, here is the important part. I brought back 8k+ up to 0. This gives the guitar that "air" that so many people mention. Without it the guitar sounds muffled and lifeless.

...liquid metal 2 stems incoming?

Edited September 17, 2013 by Garpocalypse

[Argle]

I didn't find the drums that hard to work with. Being a sampled kit they all were well recorded. If you thought this was hard, try mixing a live kit that has been recorded poorly.

[Brandon Strader]

I'm late to the party, but I wanted to give this a shot so I spent a few hours on it this afternoon.

https://app.box.com/s/qw5e9p2fabpaoc8nhl71

How did you manage to get that hiss in the intro? The master is very saturated almost like you put some kind of tape saturation on it, which may explain the hiss, eh?

The high end is very dulled in this version, and the bass is strong. With the high end more present, it might be more balanced. It also seems compressed or maybe it just seems that way because of how strong the bass is. The EQ on the bass guitar also seems weird. Almost like a resonant EQ that pushes up around 800Hz or so with a strong boost, gives it a strange tone methinks. As a whole it kinda sounds like a live recording with the microphones halfway away from the stage so that it has an interesting reverb sound. It sounds cool that it almost sounds like a "live recording" which is an art all in itself but your goal was probably not to do that?

The flute was too loud, the strings seemed ok. The piano and mandolin did get very subtractive EQ to them that killed their dynamics (like in the intro) kinda similar to what a previous person did.

So what did you use for reverb and such? Sends, or individual reverbs for each channel?

The acoustic guitar didn't sound as full during your sax part because you seemed to have it a lot quieter than it should be. It's kinda buried in the mix.

If I remember correctly the stems didn't have a cymbal track and were located only on the overheads. This might be the way most live drums are i honestly have no idea but it did make the cymbals tougher to work with.

Yes, that's how live drums are usually recorded, with one or two overhead microphones. So you should leave this overhead stereo WAV centered to not affect the natural panning that is already present in the track. For recording live drums, you'd probably need to pan each overhead mic separately, exporting this as a stereo wav was because that's how it is presented in the drum VST, all of the drum outputs were simply stereo. I think the overhead is a combination of those panned cymbals, as well as other sounds like kick, snare, that is basically "Bleeding" sounds. They bleed into that track, I suppose to keep it "realistic", it's common in drum VSTs to have that feature. The answer is to either highpass it until you barely hear the bass of the kick and snare or just use a low shelf to curve it down a lot. You don't want the reverb for the overheads acting on the bass of the kick and snare. The reverb should be highpassed to around 800Hz anyway so that it doesn't apply to sounds that low. It's basically mud.

The drums in the original were really compressed. Much more than I guessed the first time around.

I mostly use compression for make-up gain to raise the volume, the goal wasn't to really over-compress the drums but that could have been a side effect. It sounded good to me and didn't sound over-compressed, so it is what it is.

As for the guitar chords I don't know if you were talking about mine or brandon's (probably brandon's) but I can tell you what I did with them.

-Hard pan the guitars and send them to a stereo bus with it's own EQ.

-I can't remember the filter settings exactly but I think I HPF'ed around 160-180 with a Q of about 1.0. Also a slight dip around 500hz (more out of habit). Then around 2k-6k I used a shelf to make a wide dip around the vocals. Not a hard dip, only about -5db or a little less but it's a wide one. Then, here is the important part. I brought back 8k+ up to 0. This gives the guitar that "air" that so many people mention. Without it the guitar sounds muffled and lifeless.

...liquid metal 2 stems incoming?

NO LIQUID METAL 2 STEMS!!! I might post some kind of heavier stem at some point, but not that one.

Ok... I'll show you how I EQd both acoustic guitar tracks. I don't usually highpass or cut as much body as I did this time, I suppose the recording came out with a bit too much bass for my tastes. You can see how I handled that here:

I used a separate EQ for the highpass just because the built-in EQ doesn't have enough bands.

You've got your light predelay and reverb sends, good stuff...

I don't use a stereo bus, there's really no need to. I'll EQ an acoustic guitar roughly to how it should be, and save it as a preset, then for future songs I'll update the preset to match how the acoustic guitar recordings come out for that song. Then I'll just apply that EQ preset to all of the acoustic guitar tracks. So the settings on this (AcousticL) are identical to AcousticR. Maybe a stereo bus would make it easier to set the levels for all of the acoustic guitars in a song, but I'm not into that kind of thing, I like to do it all manually. I figure making the bus might be faster but I'm just not into it.

Any other questions or anything I missed... feel free to ask again... xD

I still need to explain my sends. I'll figure out the most effective way to do that soon, I hope.

[Magnetic Ether]

How did you manage to get that hiss in the intro? The master is very saturated almost like you put some kind of tape saturation on it, which may explain the hiss, eh?

Yeah, the hiss is just a saturation plugin on the mandolin with the noise knob turned up a little bit. The master itself isn't saturated though.

The high end is very dulled in this version, and the bass is strong. With the high end more present, it might be more balanced. It also seems compressed or maybe it just seems that way because of how strong the bass is. The EQ on the bass guitar also seems weird. Almost like a resonant EQ that pushes up around 800Hz or so with a strong boost, gives it a strange tone methinks. As a whole it kinda sounds like a live recording with the microphones halfway away from the stage so that it has an interesting reverb sound. It sounds cool that it almost sounds like a "live recording" which is an art all in itself but your goal was probably not to do that?

It's funny, I thought I was mixing the bass lower than I normally would have, but listening to it again, maybe it is a bit too strong. The bass is actually cut at 800hz but I did use a multiband compressor on the master to compress the bottom 200hz, so maybe that's what you're hearing? I'm not sure what you mean by resonant eq though.

I also ran it through an amp sim - I imagine you already amped it, but I thought it sounded nice so I went with it anyway . I also used a transient shaper to soften the attack a little in hopes of reducing some of the pick noise (...although that probably wasn't the right tool for the job). I actually didn't compress anything besides the drums and bass too heavily, so it might be the bass giving that perception.

The flute was too loud, the strings seemed ok. The piano and mandolin did get very subtractive EQ to them that killed their dynamics (like in the intro) kinda similar to what a previous person did.

Obviously there was a lot of eq for the intro, but I actually eased up on it a lot once the intro was over. The piano was highpassed at 200hz and had a cut at 1000 to make room for the vocals, but that's it. Looking at the EQ again, the bandwidth and gain on that cut is a bit extreme though, so I definitely should have eased up on that a little bit. The mandolin was only highpassed to get rid of the bottom end energy.

So what did you use for reverb and such? Sends, or individual reverbs for each channel?

By sends I assume you mean busses? I did both. I used a small amount of either PM's Digital Plate reverb or Reverberate CM on the majority of the tracks, as well as a little bit of additional reverb on the drum bus and the master track. Also used PSP's pianoverb for the piano. I also put a subtle delay on a couple of the backing vocal tracks so that they were panned ~70% with a softer delay panned opposite and I used a stereo expander for the strings and some of the backing vocals.

The acoustic guitar didn't sound as full during your sax part because you seemed to have it a lot quieter than it should be. It's kinda buried in the mix.

Good point, I hadn't realized how buried they were. Turning them up still doesn't make them sound nearly as powerful as yours though.

I'm still not sure how to create a more full top end for this piece. I didn't really cut the highs on anything but the bass instruments and I thought I boosted a fair bit on some of the higher ones. Is there some trick I'm missing to making things sound brighter?

[Garpocalypse]

Ok... I'll show you how I EQd both acoustic guitar tracks. I don't usually highpass or cut as much body as I did this time, I suppose the recording came out with a bit too much bass for my tastes. You can see how I handled that here:

...what the heck is that?

Really?

I know it's an acoustic guitar but that flies in the face of everything i've been learning about guitar processing this year. I'll have to try it.

...you boosted 2khz?! REALLY?!!

[Magnetic Ether]

...what the heck is that?

Really?

I know it's an acoustic guitar but that flies in the face of everything i've been learning about guitar processing this year. I'll have to try it.

...you boosted 2khz?! REALLY?!!

Haha, that was kind of my reaction as well...rules are meant to be broken I guess?

[Garpocalypse]

I'm still not sure how to create a more full top end for this piece. I didn't really cut the highs on anything but the bass instruments and I thought I boosted a fair bit on some of the higher ones. Is there some trick I'm missing to making things sound brighter?

I wish someone told me this years ago but your initial fader adjustment is the most important part of getting a good mix. I'm just now starting to get the hang of things but you don't want to listen to just the mids of every instrument when setting volumes. Which i think some people do without realizing it as this overpowers the low end and makes things muddy.

If you have a balanced initial fader-ing when mixing all parts should be audible even without any EQ'ing. Once you get that your mix will be even across the spectrum and any Eq'ing you do will be extremely sensitive.

[Brandon Strader]

...what the heck is that?

Really?

I know it's an acoustic guitar but that flies in the face of everything i've been learning about guitar processing this year. I'll have to try it.

...you boosted 2khz?! REALLY?!!

Did someone tell you to cut 2KHz on acoustic guitars? It's a good area to boost.

If you're working with HEAVY guitars, it's a good area to cut.

It really depends on how your own recording comes out, though. Depending on your mic placement and such when you record from the amp, results can vary pretty heavily. Dunno what to tell you about amp sims, though. I try to avoid those, they're not so natural and I often find them very hard to get a decent sound out of.

As for brightness of the entire song, maybe a multi-band comp, multi-band limiter, or linear-phase EQ (or all of the above!) can contribute to overall brightness of the song. I often find I get better results not pushing too far with these 3, but together, you can touch different areas without going overboard, and without sounding unnatural.

I don't know if you can tell from the picture but the volume for the acoustic rhythms was 1

The piano has a natural panning through the VST (The Giant piano VST) where the high notes are more to the right, and the low notes are to the left. I figure this is from the traditional mic'ing of pianos which is to use a microphone on each side of the open top. I don't usually fiddle with the panning of the piano, and as for EQ, I don't think you need a big cut to make way for vocals. It's probably a good idea to do a soft cut at 800Hz or so to make room for more of the meat of other instruments (bass, guitars, etc)

To soften the attack of the bass.... I posted a pic here before, but to shelf down the EQ for the high end gets rid of the click. It wasn't a picked bass, though, it was fingered bass. It's a heavy bass with very heavy strings, though, and the tone knob is turned wide open to allow a lot of the treble sound to be recorded, and then subtracted in EQ as you see fit afterwards. Ideally I could buy an acoustic bass, but eh. That's money.

[jnWake]

I guess the 2khz confusion comes from here:

.

In the Bass EQ, why did you raise the 600Hz section? That one usually sounds like crap in bass tracks haha.

[Brandon Strader]

I guess the 2khz confusion comes from here:

.

In the Bass EQ, why did you raise the 600Hz section? That one usually sounds like crap in bass tracks haha.

If there's 1 thing to take from all the things I've been saying about why and how I EQ'd stuff, it is that it varies from both song to song and sound to sound. I boosted 2KHz on the acoustic guitars because it sounded good and didn't clash with the vocals, which I feel have a good presence in the mix. The acoustic guitars are hardpanned and the lead vocals are dead center, there isn't much overlap. I don't think it sounds painful, but it would be just as easy to turn that boost off and compensate by maybe raising the acoustic guitars volume by 0.5db in my mix. Maybe not even boost the volume, depends on how it sounds. Listening to how it sounds is more important than following some "standard rule for all songs". Maybe the 2KHz of the guitar wasn't that strong to begin with? All recorded instruments will sound different depending on a huge range of factors, from mic placement, to string quality, and beyond. These strings were not completely fresh but they were still kinda vibrant. I tend to EQ acoustic guitars differently based on how old the strings are because you DO get a different tone from them as time goes on.

As for the boost on 630Hz on the bass, again, it's based on how it sounded to me. Different basses will need different EQ. You can't apply the same settings for every song or every type of bass. Not even the SAME BASS played for a different song. Things will always tend to sit differently based on all of the sounds you use in the song.

I haven't experienced any 'pain' listening to My Choice over and over, and it sounds very cohesive when played on a big system with the subwoofer cranked to max. Experiment with EQ, don't be taking notes like "I should always boost my bass at 630Hz", that is NOT what I am setting out to say here. That's just what worked for this song, and it will be different in other songs.

[jnWake]

I'm not trying to find rules in your EQs, I just thought it was curious that you raised the 600Hz. Maybe you did it because of a particular reason that could be useful sometime (other than "it sounds good"), but apparently this wasn't the case

[Brandon Strader]

Well I don't know the best way to explain it, I don't want to misinform. I'm not home to double check what difference it makes, but I know that general area (usually around 800Hz or thereabouts) is attack / honk or something of that nature, so I would have boosted there to emphasize the fingering of the bass a bit better.

[Brandon Strader]

I was wondering why nobody had been commenting or trying this anymore...

[Darangen]

I just now saw this.

I'll give it a go, seems fun.

[Brandon Strader]

I just now saw this.

I'll give it a go, seems fun.

Sweet, excited to hear it!

I'm probably going to release some metal stems too, with heavy guitars, growls, screams, clean vocals, it'll be nice and good for practice. I need to record the lead vocals for the song first though, been on the shelf for a while now.

[Eino Keskitalo]

I think this is a good idea and a cool thing to do. I haven't had the time to check it out though.

--Eino

[Darangen]

I know it's been a while, but I finally got around to this.

https://www.dropbox.com/s/f0ha9x0ecy3mt8k/mychoice.mp3?dl=0

I haven't listened to any of the other mixes of it yet so I'm not sure how it measures up, I just mixed it how I would've imagined it should sound.

I went for a ambient feel with a decent dose of reverb and delay. I let the percussion take a back seat so the instrumentation could have a little more attention since there wasn't anything incredibly complicated in the drums. I felt the acoustic guitar strums could carry a lot of the weight of the rhythm.

[Brandon Strader]

Dang, that was awesome. Did you make the piano mono or did I export it like that? :3 It's been so long I honestly don't remember. I really like this version, takes a different approach than what I've heard. Digging the increased reverb, and the vocals even sounded pretty good the way they were treated here, a bit high in the mix maybe. From reading your text I was expecting the drums to be muted completely. I like the balance, it's pretty interesting.

[Darangen]

The piano was stereo but in one track. I made it a little more mono with a stereo imager because I wanted the acoustic guitars to surround it.

[anterroir]

Interesting track -> rough headphone mix

[Brandon Strader]

It's been years but the files are still active on dropbox.

I am still learning, and always trying to do things a little differently for the better, or at the very least to get a different sound. For one thing I want to experiment in 2016 with making my mastering chain SMALLER. It has a lot of parts to it that aren't as necessary as I think. You could really get by with a simple linear phase EQ and limiter.... yeah -- that simple. But I want to also make it possible to put a compressor in there somewhere -- maybe a multi-band compressor. Nothing too harsh, but for applications where it would be useful, I'd have it, even if I have it turned off most of the time. 

How have the rest of you guys been updating and adjusting your mastering chains? What new discoveries have you made with EQing? After all this time I'm a little surprised at how I EQd those acoustic guitars, some things I've forgotten or lost along the way. High passing was probably a great idea, but I have not done that on acoustic guitars last year. I probably should. Body is good to have, and some low end, but not at the risk of having too much. You've always got the bass guitar to fill in the low frequencies anyway.

I started using a different EQ in Cubase, I think it's called EQ30 or something similar, it has a lot more bands and more control, but seems nice and transparent. 

Also just to add, I want to experiment with using a limiter on heavy guitars. I know it is frowned upon at OCR to overcompress your guitar tracks, especially rhythms, but in the professional metal field using a limiter on rhythms is 1 of the single most common things. I know for a fact, from having talked to Vladimir from Mirrorthrone, that he uses a compressor on his rhythms or at least he did during the recording of Gangrene album which was also years ago. But I see the value in limiting more than compressing to get that nice guitar brick rather than a sausage

[timaeus222]

Yeah, I just use one EQ instance (minor high pass near 28 Hz) and a limiter, usually, and occasionally, if I want a punchy final result and a little more cohesiveness ('gluing' together the instruments), a compressor for parallel compression. After I do all my mixing, there's not much I ever need to do on the Master track or post-render. In fact, I program my synth sounds so that they're clean (few overboosts or hollowness, if any) out-of-the-box, to minimize the amount of EQ I need to do to make them sound more "even" across multiple audio systems.

These days, much of my EQ is fairly simple (low shelf below 200 Hz, a little mid-scooping, maybe a little low pass near 18000 Hz, etc), and then sometimes I'll take another pass through and do notch EQs and EQ band gain automation. The better the frequency distribution your sounds start out with, the less you need to EQ.

I don't remember off the top of my head, but most of my electric guitar processing probably has a limiter or soft-knee compressor somewhere, either in the amp sim or within Shreddage 2X itself, just to keep the peaks controlled.

[Garpocalypse]

3 hours ago, Brandon Strader said:

 

How have the rest of you guys been updating and adjusting your mastering chains? What new discoveries have you made with EQing? After all this time I'm a little surprised at how I EQd those acoustic guitars, some things I've forgotten or lost along the way. High passing was probably a great idea, but I have not done that on acoustic guitars last year. I probably should. Body is good to have, and some low end, but not at the risk of having too much. You've always got the bass guitar to fill in the low frequencies anyway.

 

I know i'm not as experienced as you guys are but my mastering chain is just a Multiband compressor used to give a slight pump to the lower and upper mids, a console saturator, and a booster/limiter  (Boost11 it's called). I try to avoid EQ's on the master but if I have one it's usually a slight low shelf at -2db or something slight like that to balance out some of the lows.  Blending Reverb and delay is taken care of via sends on the faders in almost all of the stuff i've done lately.  

 

3 hours ago, Brandon Strader said:

 

Also just to add, I want to experiment with using a limiter on heavy guitars. I know it is frowned upon at OCR to overcompress your guitar tracks, especially rhythms, but in the professional metal field using a limiter on rhythms is 1 of the single most common things. I know for a fact, from having talked to Vladimir from Mirrorthrone, that he uses a compressor on his rhythms or at least he did during the recording of Gangrene album which was also years ago. But I see the value in limiting more than compressing to get that nice guitar brick rather than a sausage

I find that my metal processing considerations tend to change every few months but one thing i am pretty happy with is the sound of my rhythm guitars.  99% of metal rhythm guitars sound horrible imo and I don't see the point of a limiter on the rhythms especially if you are talking high-gain. Of course I follow power metal, which maintains a more aurally pleasing rhythm guitar distortion, much closer than metalcore/nu-metal/industrial where compressing guitars an insane amount is common.  Since most metal mixing uses the rhythm guitars as the base of the mix in the first place, what exactly would you be accomplishing with a limiter?

A much better idea methinks would be to put a multiband on the guitarbus or post amp-sim to give it a slight pump in the upper and lower mids.  This makes the rhythm guitar a bit more subtly dynamic.