How do I save this abomination

[Stephenar19]

I'm still fairly new to home recording, I've been following a lot of tutorials and guides online to learn the basics of recording and mixing for the last 10 months or so, and some of the tracks I've made so far have turned out pretty good, and some....not so good. Specifically, I think I might have fallen into the trap of sinking all of my time into a dumpster fire, but I've been working on one particular track for way longer than I care to admit and used it as a sandbox to try out new plugins, automation, re-amping, different recording techniques, amp sims vs. my axe-fx, etc. For all the time I've put into it though, the end result is truly pretty awful.

The problem is, I don't actually know how to fix what makes it so bad. The guitars sound staticky and thin to me, the bass either sits completely unnoticeable, or becomes horribly intrusive if I raise the volume in the mix, and the drums all sound very disjointed. The composition is certainly weird - I was going for a sci-fi/horror video game soundtrack aesthetic - but the end result is something that sounds so far from a professional mix that I want to just throw it in the trash never to be seen again. Unfortunately though, it's part of a 4-part project that I'm working on (and the other 3 aren't too bad imo) so I'd really like to save it if I can.

I do think that part of the problem might just be poor guitar performance when tracking - but I think there's also a lot more to be corrected in it as well. To the more experienced folks out there - what really stands out to you and what would you focus on fixing?

 

[Stephenar19]

For context I'm doing all work in Reaper with almost entirely stock plugins (save for amp sims, submission audio bass, and EZDrummer)

Guitars were tracked through an Axe-FX II, which I'm using as my interface. Guitars were originally tracked with a self-made tone but I eventually gave up on that and put the DI through Neural DSP Nolly

All monitoring was done either through a set of Yamaha HS7's or through headphones (DT 990s) - and I've been rendering the tracks and listening to them over on various speakers - car stereo, bluetooth speaker, earbuds, phone, you name it

 

[TedEH]

Take everything I say with a grain of salt because I'm not a pro, but I think I'm hearing most of what you described.

What stands out mostly to me is that everything is kind of disjoint - nothing is glued to anything else, nothing sounds like it's in the same space, and occasionally there's very dramatic balance or volume shifts. The most dramatic imbalance is the drums vs. the guitars. The drums are distant and spacey, but the guitars are dry and close and present and louder than everything else, so you get that sort of effect of guitars sitting loudly on top of the rest of everything else. I have a suspicion that you have a preset for your drums, and that preset just doesn't match the way you've processed everything else.

I think you need to step back and thing about each instrument's place and what it should be doing in relation to the whole.

A wall of general advice that might or might not be helpful:

Usually, I tend to think of the drums and bass as the backbone of everything, so if you start just from there, and strip away anything fancy - automation, effects, etc.
- The drums don't sound bad at the beginning when they're just starting, but they're a bit distractingly wide, and there's some kind of reverb or delay or something, or maybe a compressor pumping or something that's making the hats sound weird.
- The bass starts off fine, but it gets buried as soon as something plays over it. If you want it to stay audible, you need to pick a range that "belongs" to the bass and not let anything else step on that.
- On top of giving the bass some ownership of a certain range, bass as a whole tends to stack up, so be prepared to take low end away from anything that isn't a bass instrument. Some people hate taking bass away from guitars. Suck up the guitarist ego, and do it anyway.

Every element you add should be bussed together and treated as a unit. The bass? One unit for all bass tracks. Drums? Put it on a bus, so you can compress/eq them together. Guitars? Sum them all to a bus and handle them together. Reaper automatically routes things by folder structure, so organize your tracks by folder, and work on groups through the parent tracks.
For example -> for raw drums sounds (I know you have samples, but this is just an exmaple), I'd have each drum on a track, with a bit of processing each. Then all the toms get grouped and compressed together a little bit. Then all the shells (toms/snare/kick) grouped together and compressed a little bit. Then those are grouped together with overheads and the room mic and the whole thing gets a bit of compression. Putting compression at each layer "glues" keeps each element consistent, but also glues the groups together. But you have to be careful with this because a lot of compressors means a lot of places things can go wrong, and a lot of investigating if you have to troubleshoot something. Point being -> use folders. Organize your tracks into folders. Folders are your friends.

As you introduce each element (probably add guitar after you have drum/bass ready for it):
- Think about the frequency range you care about. Guitars don't need much low end at all, despite what some people might intuit. Guitars can also roll off high end sooner than you'd expect. You can low-pass as low as 4k or 5k and still sound fine, depending on the guitar sound and everything else around it.
- Think about how far away the thing should sound. Something that is close will sound dry and have more high and low end. Something that is far will lose a bit of low end, but lose a lot of high end. The balance of reverb to dry sound also is a marker of distance.
- Think about the space this is in. Are you trying to convincingly sound like this thing is in a room? What does a room do to a sound? is this thing in the same room as the drums? If two things are meant to sound like they're in the same space, they should have comparable dynamics, balance, amount of reverb, etc. Dark verb-y drums next to bright dry guitars will sound disjoint like what you have now.
- Think about what else needs to compete with this instrument. If you put a mic in front of a Recto, it's going to compete with your bass. If you don't use a lo-pass and keep a dry guitar sound, it will compete with pretty much all "high frequency" stuff. You have to make cuts to make stuff fit together.

Past that, just go a step at a time, and don't do anything without a reason. Change something, listen, evaluate, make another change based on your evaluation, etc. Don't just follow step "because you're supposed to". Work with purpose.

Don't mix with the aim of making anything "loud". Loudness is a mastering step. If you aim for making everything is loud and present as possible, then each element will just be fighting every other element forever. Use the mix to tetris and glue everything together, then mastering will make it loud later.

Probably leave automation until after you're satisfied with the overall mix. There's a good change it'll just confuse things if you throw it in too early.

Hopefully something in there will be useful.

 

[lost_horizon]

Abomination? Name of the horror movie?

How about this, take it back to midi, everything, including the guitars, transcribe it. If something sounds really off it's a tonality thing, if it sounds off after you play/record it, it's a tuning/recording thing. If it sounds good just notes, it will sound good with the guitar playing it.

Sometimes with difficult parts I split them into just 4 bars, do 5 takes, keep what works best.

There is a disjointed sound between the drums and guitars. You got two options. Transcribe the guitars and use them as a base for the drums or take the drums and write guitar parts that fit.

Keep the bass simple, put a power chord that works with the bass, if you want to add some more colour add the 3rd. If you want to move it around do an octave.

Drum sounds? Sounds super dry, at the start it sounds like midi drums. Try a new preset, the dynamics are good. There are little bits of the drums that hit off beat that the other bits need to mirror somehow. Can you tune the drums? Pick a low or particular drum note to hit. Like the bit at 2:20 does.

Bass something like a rickenbacker etc that has a full but spacey sound, leaving some room for the guitar.

The lead guitar is loud but not in time with a particular hit. The guys from Polyphia write all their stuff out in midi with synth voice and then attempt to play it. Try and build something in the lead. Is the guitar effects synced to the recording? Use the track tempo in all timed effects.

Is the violin part in the right note range? Here is where it might not pay to be playing the exact same notes, you might want the bass and violin to be more simple and provide a structure the lead and rhythm can go over. Just like any other instrument samples can have a good and bad part of the range.

Dynamics of maybe a clean part? Some open chords can really flesh out a section.

There are plenty of cool atmospheric packs and vst's out there. There is one called the Joker from the Joker movie available from Rast sound.

Joker Tones - Rast Sound

Inspired by the story and character of Joker, we couldn’t resist to produce this free library offering dark, deep, mad and nostalgic instruments and sounds for Kontakt & WAV. “Thank you for the unusual library! There are some very interesting and usable sounds in there.” FEATURES + 370 MB for...

rastsound.com

 

[Stephenar19]

Awesome, thank you both for the listens and extensive feedback. I’m at work now, but based on your advice, when I get the chance I’ll strip this project back of all of the automation and effects and try to rebuild it from the ground up.

Some things that you noted that stuck out to me:

1 – I never even really considered what kind of space I was trying to emulate with the song, so I think lost_horizon hit the nail on the head with pointing out why the drums sound so tonally disjointed from the guitars. I’ll start with the drums and what kind of tone they convey and then add the bass and then try to build around them and match the guitars

2 – TedEH pointed out that there was something pumping on the hats. This was me playing around with sidechain compression, but to be honest I had it at a level where I couldn’t notice it anymore so I just left it on. I think I need to work on my ears and recognizing compression – I watch a lot of tutorials where they’ll A/B two settings and I’d say 8/10 times I don’t actually notice what’s changing.

3 – The guitars definitely need to be glued together – I don’t know what I was trying to achieve here lol. I have like 3 different sim settings and each part is EQ’d differently. The fix there is pretty straightforward – I just need to put the time in and make them all consistent. Part of the problem is that I keep going back and forth between ‘man I hate this tone so much’ and ‘I shouldn’t be focusing my time on the tone, if I have a good mix it shouldn’t matter too much’.

4 – I think I’m going to strip out all of the EQ settings and automation and focus on just making simple adjustments only as needed once a basic mix has been established. I was trying to EQ and adjust as I went and I think that was a big part of my problem. This also ties into the master bussing for the tracks. I started this project before I learned about that, so I started bussing together tracks which already had fx on some tracks but not others, leading to a disorganized mess.

Again, thanks for taking the time to type this all out, I really appreciate it!

 

[TedEH]

The guitars definitely need to be glued together

It's worth remembering that the goal is not just to glue the guitars to themselves, but to glue to the guitars to everything else. That's what bussing and compression can do. The goal of a mix is the tetris and glue - that's not the artistic stage. Crafting sounds comes before the mix. The then mix fits them together.

Remember that there's always a why behind a change, and there's more than one way to do everything.
Consider the idea of a bass guitar and a bass drum. They might be competing for space. They might not be depending on how each is tuned and played. But if they are, a lot of tutorials will tell you to fix this with a sidechain compressor. You can do that if you want, but you don't have to. Sidechain processing is harder to pull off than just plain compression. You could also use some creative EQ so that emphasis for each is in a different range, and then put a compressor somewhere higher in the chain that covers both of them. Just compress them together. If you put two things under a compressor, and one becomes temporarily louder, the compressor will have the effect of ducking both of them. That ducking effect is your glue.

 

[Drew]

I do think that part of the problem might just be poor guitar performance when tracking - but I think there's also a lot more to be corrected in it as well. To the more experienced folks out there - what really stands out to you and what would you focus on fixing?

two comments, and I'll preface this by saying I'm at work so I haven;t listened.

If it's poor guitar performance, the best thing you can do is retrack them. Nothing will make a mix sound "tighter" than really tight performances.

Next, shut every single effect off the mix, and just set levels and pan. Walk away for a while, then come back and listen, and spend some time thinking about what you're hearing. Try to identify the things that sound like "problems" to you, and when you do, then try to think about what you can do to fix them (hopefully while mixing, but a perfectly valid conclusion here could be the problem is in the performance or in the arrangement and going back and retracking some more could be the answer). Once you have an idea what you need to address, then and only then start adding FX.

It's easy to go down this rabbit hole where "well, all the pros say you need to add this to your gutiars, and this to your bass, and this to your master," etc, but I'm a firm believer that whenever you reach for a plugin it should be because there's something you're hearing that you expect that plugin to fix, and sometimes just starting from scratch with a clean slate can be the best thing you can do.

 

[hensh!n]

When mixing or doing anything creative really, there are always going to be two sides. One is subjective (e.g., the sound is "harsh"), meaning artistic in the sense that you're intentionally doing something a particular way. The other is objective (e.g., the performance isn't in tune), meaning something is objectively inappropriate for what would be considered acceptable. I try to focus on objective things, since art at it's core is and should remain subjective.

The first thing that stands out is the drums. For me, they aren't loud enough. I tend to view drums as the backbone of everything. Get the drums right first, then drop the bass in to lock it with the drums. With bass it's easy to get carried away as even bass players will tell you it isn't necessary to have a lot of bass in a track unless it's an EDM track. Another thing is the guitar, I won't tell you it sounds bad (it's subjective) but I think it doesn't quite fit the mix. Also, each guitar that is coming in sounds contrastingly different, which in a certain context may work. However, here I felt like it was distracting to the point where I was focused on how each guitar sounded so different. With every song you're guiding the listener to your vision, I think ultimately you sound a bit conflicted and confused. Whenever I feel that way I find that stepping away is usually best. If I revisit a track, I try to make fast decisions and changes. Pro mixes usually mix on intuition because they know that time is against them. Not necessarily a deadline, but they'll use their initial spark or objectivity and begin to make subjective changes that aren't needed.

Another piece of advice I can offer is don't change or do something for the sake of it. Use your ears. Listen if the drums aren't loud enough, if the bass is too loud, or if the guitar isn't quite suiting the feel of the song you're trying to create. It's best to not create problems that don't exist and if possible, to have a basic understanding of what you're doing and why you're doing it. But don't worry, these issues have happened to everyone at some point.

If you follow a lot of basic fundamentals and hone your craft, you'll be the majority of the way there. I would suggest spending more time on fundamentals like timing (e.g., drum fills) rather than which compressor to slap on a drum bus or something like that. Anyways, I hope this helps.

 

[TedEH]

The first thing that stands out is the drums. For me, they aren't loud enough.

I know there's already a lot to take in, in this thread, but another point worth making:
Everything in audio is relative. Sometimes if something "is too quiet", you could equally say that something else is too loud. The correction is to make them even, relative to eachother. EQ is the same. Sometimes you might hear something and think "this is too dark, it doesn't have enough high frequency", when in fact it has plenty of high frequency, but even more low frequency. So you get brightness not by boosting highs, but by cutting lows. Much of the goal of mixing is to bring all things in balance with eachother.

This is, IMO, why people say most EQ you do should be cutting first. Cutting low end, in particular, gives you headroom back and lets all the competing elements sit side-by-side without competing as much.

 

[hensh!n]

I know there's already a lot to take in, in this thread, but another point worth making:
Everything in audio is relative. Sometimes if something "is too quiet", you could equally say that something else is too loud. The correction is to make them even, relative to eachother. EQ is the same. Sometimes you might hear something and think "this is too dark, it doesn't have enough high frequency", when in fact it has plenty of high frequency, but even more low frequency. So you get brightness not by boosting highs, but by cutting lows. Much of the goal of mixing is to bring all things in balance with eachother.

This is, IMO, why people say most EQ you do should be cutting first. Cutting low end, in particular, gives you headroom back and lets all the competing elements sit side-by-side without competing as much.

Yeah that is a good point. It's why we should have a working methodology or workflow. You don't want to consistently be changing levels and going back and forth on decisions. A methodology should allow you to design a structure for a mix that is relatively balanced. Or relatively balanced enough to begin working effectively and efficiently.

I have templates and my own presets to give me a rough starting point every time I'm building something.

 

[Stephenar19]

Yeah that is a good point. It's why we should have a working methodology or workflow. You don't want to consistently be changing levels and going back and forth on decisions. A methodology should allow you to design a structure for a mix that is relatively balanced. Or relatively balanced enough to begin working effectively and efficiently.

I have templates and my own presets to give me a rough starting point every time I'm building something.

This is the exact trap that I find myself falling into. I've also been following frightbox on youtube and the importance of a standard workflow seems to be one of his main points that he keeps coming back to. I'm finding that establishing that workflow is much easier said than done though, so I'm still trying to feel my way around until I figure out a process that works well for me.

 

[cthsqd]

I'd say that looking for the perfect result at the first try is meant to fail. Simply, because it comes slowly with the experience, not only in understanding the daw/workflow/mixing but also in our perception. Simply, some of our first songs must be worse so the next ones can be better, and there is no way around it.
Certainly looking for knowledge (through feedback) is the right way in the process.

 

[Stephenar19]

Update - I took the suggestions you guys gave me and tried to retake some of the guitar parts and remix the track. It improved a little bit, but I still think it's a pretty rough job.

Anyway, I decided it was time to take my learnings and move on to a new project and so I decided to call this "EP" finished:



I hear some really awesome stuff posted to SSO and I realize this is nowhere near the caliber of what a lot of other folks are sharing, but I'm a noob and open to any and all critiques/suggestions on how to improve my recording/mixing/mastering. Seriously, please rip it apart