Recording bass and guitar for YouTube
October 17, 2019
About a year ago, I started learning about how to make my own bass and guitar videos to publish on YouTube, whether it’s songs covers or original content.
While I’m far from having figured everything out, this article is a work in progress on the things that I’ve learnt so far on the topic, that might be useful for beginners who want to do the same thing, and especially for myself to avoid repeating the same mistakes for my next recordings!
This article is somewhat specific to the gear and tools I use, but I’m pretty sure the general ideas can be applied to many different setups.
To start with, here’s the gear and tools I’m using.
Make sure to have enough storage on your SD card for the video you plan to record… it’s annoying to run out of space and have a take cut in the middle just because you forgot to format the SD card before recording and it contained 42 GB of videos of an old video that you already transferred anyways.
Make sure you have enough battery when you start recording, ideally charge it to 100% just before. If you can plug your camera while recording to avoid running on battery, it’s even better. It’s pretty frustrating to run out of battery in a middle of a take and have to redo it.
That’s probably the only mistake that I make quite regularly when it’s about recording music videos; I forget to focus, or I focus on some object in the background of where I’ll be standing while I’m behind the camera, and then I forget to focus on myself when I’m actually playing.
Too many times I’ve ended up with a great audio take but unusable video because the focus point was not on the subject (me), which is pretty frustrating.
The solution will depend on your gear, but as far as I’m concerned my camera have a Wi-Fi mode where I can remote control it from my phone, so I just need to remember to do that and adjust the focus point from my phone before I start recording.
Another option is to have someone else behind the camera to take care of this kind of details. That can also make your life much easier for framing, as you won’t need to go back and forth between the camera and the spot where you’ll be standing in the video to adjust the framing. It usually takes me 4 to 5 roundtrips like this (including taking my bass and putting it down again every time) to get a decent framing and focal length.
When I’m satisfied with a take, I stop the recording and take note of the time and the number of the take in Logic so that I can easily match up the take with the associated video when editing.
If your audio interface jack input have a Hi-Z switch (stands for high impedance), turn it on, as it’s gonna give you a clearer sound. On mine, only one of the 2 jack inputs have a Hi-Z switch, so I make sure to only use this one when I’m recording bass and guitar.
To calibrate the input sensitivity, most interfaces feature a LED that indicates clipping. I usually try and play something louder that I will need, and make sure that the input sensitivity is at a point where only the “too loud” stuff results in clipping, meaning that when I play normally I’ll have optimal recording level.
I’ve found though that when slapping the bass, I had to turn the sensitivity very low to avoid clipping, which then made it harder for me to have the bass sound loud. It seems to me that the clipping happens on some kind of extra “noise” when I slap but that it’s not directly clipping the actual bass sound, so especially for slap, I allow the clipping LED to light up a lot, as long as the overall bass sound doesn’t sound distorted. Basically I end up turning up the sensitivity as much as possible regardless of the clipping LED as long as it still sounds good.
Whether it’s bass or guitar, you’ll get a significant tone improvement by changing your strings before recording. New strings just got that extra clearness in the tone, that extra “zing” that gives some colour to the sound, and it usually makes for a higher quality recording.
However I record pretty much every week, and that new string tone goes away in a couple days for me, so I just accept that most of the time I won’t have an optimal tone, as I don’t really want to change both my guitar and bass strings every single week.
One trick I’ve found brilliant for bass though, consists in loosening the strings and “slapping the shit out of them” before tuning it again and playing, as shown in that video.
This turned out to work really well for me, even though I can’t get 100% back the new string tone, it already sounds much better than the dead string tone I get after a week or so. However I need to do it every time before playing, otherwise it’s a matter of hours for the tone to go dead again.
There seems to be mixed opinions online on how coated strings could help with that, so I haven’t tried that just yet. It seems also that wiping the strings with some kind of cloth after playing should help preserve them a bit longer, and maybe also applying some kind of product like Fast Fret. I’ll update this post if I end up trying some of those solutions.
This kind of problem seems to be very “personal” though, in the sense that some people have more sweaty hands than others, and more or less acidic sweat as well, and that seems to affect a lot the lifetime of the strings. It looks like I’m of the kind that destroys strings in a couple hours of playing, but if you’re lucky enough you might be able to keep your new string tone for months without having to think about it.
When you create your project, make sure you are in 24-bit and 48 kHz mode, as this is the audio settings that YouTube expects for high quality videos.
By default Logic is set to 44.1 kHz (which would be good if we wanted to publish the music on CDs), but 48 kHz seems to be the standard for anything digital and especially audio to be used in a video, so that’s what we want here.
It looks like the default bit depth of Logic is already 24-bit but just make sure it’s the case for you as well.
Logic defaults to normalizing audio upon bounce. From my understanding, this means that it will raise or lower the overall audio level so that the highest peak is the highest possible level that doesn’t result in distortion (0 dB).
Having this enabled means that what you will bounce will potentially be different from what you’ve been listening to while mixing. I want my output to be exactly what I heard while mixing, so I keep normalizing disabled in the bounce settings.
However raising the overall level to have the peaks close to 0 dB, and also not having the master level ever exceed 0 dB is actually useful, so we have to take care of that directly in the mix (this way we can hear what the final result will sound like while mixing without any surprises).
For this, I use Logic’s adaptive limiter as the last effect of the output track, as recommended by this article, with the settings they recommend: Gain of 0 dB, Out Ceiling of -0.1 dB, 20 ms Lookahead and Remove DC Offset turned on.
I also turn on True Peak Detection (this article convinced me of that) as without it it would mean that “your mix might still be clipping without you knowing it”.
Since I was (and quite still am) quite confused about using the adaptive limiter as opposed to just the limiter, I found those two threads that explain the differences and what might be “better”. Seems that the answer is usually another plugin, but the second best answers seems to be in favor of the adaptive limiter, so I’ll go with that as it sounds decent to me.
I’ve struggled a lot (and still kind of am struggling) with the bass sounding loud enough in the mix compared to other parts. Even with compression and adjusting the gain/volume so that the bass is always close to 0 dB, it just sounds far behind other instruments, and putting it louder results in clipping, or if relying on Logic’s audio normalization, results in lowering significantly every other part and having the whole mix sound quiet.
The best I’ve found so far (which is probably not the right thing to do, please contact me if you actually know how to do that properly), is adding a limiter as the last plugin of the bass track pipeline, and boosting the gain (I usually boost it by 12 dB or so for it to sound loud enough compared to other parts).
I put the loudness meter plugin after the adaptive limiter to monitor the perceived loudness of the mix.
Looks like there’s lot to say about the audio loudness online, and I found this 3 parts article while researching on the topic that seems to explain it pretty well.
Looks like for online streaming we should aim for 16 LUFS, so that’s what I monitor for.
When doing bass covers, I usually try to dim the bass of the original track so that it doesn’t conflict with my cover. For this, I usually add a channel equalization on the original track where I set everything below 250 Hz to -24 dB.
This usually works pretty well, but it also often results in removing the drum kick, so sometimes depending on the song, I try to compensate up a smaller frequency range between 60 and 70 dB to bring it back. It won’t sound as good as the original track obviously, but will leave more room for the sound of the actual bass cover which is the most important part of a… bass cover.
Also depending on the way the original track is mastered (especially when they’re brickwalled), I usually lower the original track by about 6 dB, to allow more room for the parts I’m recording (since it’s the whole point of a cover).
I sometimes end up doing a lot of takes for the same cover or parts of it, and this ends up taking some space. When I’m fully done with arranging the takes and editing, I remove all the takes that were left unused. Then, in the project audio files panel, you can Select Unused in the Edit menu and delete those files.
This won’t actually delete the files from the disk; after doing that, go in the main window in File, Project Management and Clean Up. This will prompt you to actually move the unused audio files to the trash.
While not directly related to bass and guitar covers, here’s the little tips I needed with Premiere for doing those videos.
If you have multiple video takes for the same project and you want to apply the same color correction and creative effects to all of them, add an adjustment layer in the project panel and drag it in the timeline above all the videos. Then you can modify the colors on the adjustment layer and it will take effect on everything below it, instead of having to apply the same settings on each video individually.
As with Logic, when done editing, you can do Edit, Remove Unused to remove all the unused media from the project panel.
After doing so, I manually compare the files I have on disk with the remaining files in the project panel and I remove the ones that are not used anymore. I’m not aware of an automated solution for this, but for this kind of project it’s usually pretty quick.
YouTube will auto generate 3 cover pictures from the video, but from my experience they’re rarely the best and I’m better off picking a custom frame from the video (using Premiere’s export frame feature).
The main thing is to do that as early as possible, especially before actually publishing the video, otherwise if you start sharing the video to other platforms, e.g. Facebook, it will cache for a very long time the cover picture and even the “sharing debugger” will not do anything about the cover itself for at least a couple days after you change it.
Also when sharing on Facebook, you might want to check this article to make sure the mobile layout of your YouTube post is the full width one.