Preramble:
I’ve ended up with a decent collection of keyboards over the years of all various types, as this hobby has always been my second go to behind audio, and now I finally feel like I’ve reached the point of satisfaction in this hobby where I just enjoy typing and using what I have every day. Having tried all of what I really want to try, gotten myself to a level of experience and knowledge on the hobby that I feel mostly satisfied with, and becoming disinterested in the community and new releases, I am now to the point of wanting to consolidate down to what I truly love. Before I let things go, I thought it would be nice to try and capture the sound and thoughts I had of each board before I decided on what to do with it, along with forcing myself to take a closer look at each board and what value it truly brings me, and work to extract as much value out of them as I can, or decide to sell if I don’t see the value after doing that. I had already been recording my boards for years and years, but only did I really decide to do it seriously starting at the beginning of this year when the hobby truly slowed down for me (although it has always been up and down over time to be fair lol). When I got over my fears of doing things publicly enough to start a channel, I started taking recording boards as seriously as I could, to try my best to capture the moment as faithfully as I could. I never really paid much mind to the sound of a board if it felt really nice, it was always a second priority, and now that I was planning to make “content” in a sense, I was curious how most other people had been doing it, as I had only listened to things on and off over the years. What I found mostly just made me disappointed and depressed lol, not everything, but most of what I heard wouldn’t cut it for my own standards, and that made me really stop and think what the actual best way to go about this would be. There really is no one-size-fits-all best way when it comes to recording, I learned that long ago doing production work as a side job and as a hobbyist, so I started from the ground up on experimenting and refining my setup to best record the boards I had, hoping to meet my own standards. The following rambling will be a steam of consciousness on what I feel like I’ve learned over the years for recording keyboards specifically and what I think has or will lead to the best results, for my own preferences, based on only my own experiences, and along with some personal thoughts on the hobby itself that might be relevant for this topic.
This will be only relating to audio recordings of keyboards, I can’t speak to video lol. It’s also not about the keyboard and build itself, this is only about recording what you have and the nuances surrounding that, not how you go about creating a good sounding board. I don’t really like to make writeups like this because I don’t really feel qualified enough to do so, nor do I feel that I cover things in a comprehensible/useful manner, but I wanted to collect my own thoughts and experiences on what I believe creates a good sound demo of a keyboard, and the aspects that go into that. I’m hoping that forcing myself to organize and further analyze my own experiences will lead to improvements in my own process, but I also feel as though it’s a sorely neglected aspect within the hobby that most people just don’t think about beyond a surface level. I’m no expert or professional, just a hobbyist with some experience in audio (in both playback and recording), along with being pretty lazy, so take things with a grain of salt as my approach to things may feel a bit unorthodox, obsessive, and/or pedantic. It’s also something highly specific to each person, their setup, environment, goals, and ears; please do not take this as a “how to” and instead just an example of one person’s experiences on how something was done to their own taste. I will include some general advice that I think may be a helpful starting point, but this is not a step by step guide to follow (if that’s even possible with my shitty writing) as that’s not how you get a good result. I also may end up oversimplifying or generalizing some aspects for simplicity but I will try to avoid that. I have a bad habit of wanting to place a thousand disclaimers before everything I write, I’ll stop myself here and get on with it lol.
Figuring Out Goals:
I think it’s important to start off by determining the main reason why you want to even record a keyboard in the first place. Are you recording to showcase a final build? Create entertainment/content for others? Document the current state before a rebuild? Review a product? Compare/contrast against something else? Just cause? All valid reasons to go about it, although depending on the reason, you may end up changing how you go about actually doing it. Personally, I record my boards to create a snapshot in time of a build for how it sounds and what I think about it, occasionally I’ll also record pre/post rebuild to compare against what it was and what it is now. In these cases, I’m most concerned about recording something in a way that most accurately represents how the build sounds to my ear when actually typing on and using the keyboard in daily use in my setup. If I were reviewing something, I would place more emphasis on replicability and reliability for transparency/fairness and to allow people to cross check things at home. If I’m aiming to specifically compare, I would focus on doing two comparisons, one with neutral conditions that don’t overly favor one board over the other, and then another that are closer to ideal conditions for both boards (within reason). If I was making entertainment/casual content, maybe I would be a bit more generous in terms of sweetening up the final sound, ending up with a more ethereal/unrealistic but more “pleasing” and agreeable final result than reality. If I’m doing it just cause, then I wouldn’t really give a shit about most of this and just go for it and upload without a thought in my head (would assume those reading this don’t fall into this category lol). I believe some shared commonality behind all of these goals is the desire to have a “good” (to your own taste) recording of what you’ve created and/or experienced, and the desire to convey that experience to others, and I feel that putting in the appropriate amount of effort, thought, and care into crafting an appropriate recording will increase how well that’s actually conveyed to the listener.
Equally important is making sure to be clear to convey your goals, and the context behind how you achieve them to who you’re sharing them with, so they can fully understand and appreciate what you’re going for, and the path you took to get there. I think the most meaningful demos are the ones that go beyond just the actual result and mention both the reasoning and the method, as that gives people a much deeper understanding of what you’re actually doing, and results in a better final product as well IMO. That’s just me though, do what you like, I just personally find more value in recordings with more information than not. For myself, on each recording I’ve uploaded, I make sure to note the conditions, equipment, comments/concerns, multiple references, and explanations behind my recordings in short form, and have elaborated further in my schizo rambling current setup videos for more information. I’m not fully happy with the level of detail I’m limited to within a description or comment, so having a document like this helps to solve a bit of that as well lol. In a lot of content I’ve seen posted, there’s typically a stark lack of information about any information, goals, and context, and that just kinda diminishes the potential value I’d personally get out of a demo, even if the demo is good itself, so I wish it was more common to see. This isn’t to try and suggest most people waste as much time as me on it, but it would be nice to see at least something, even if it might seem trivial or obvious to an experienced hobbyist, it is important to consider that not everyone in this hobby has the same level of experience and insight. Your context behind the recording then offers new insight into what’s presented, and also allows the listener to get a better picture of what something actually sounds like.
I don’t want to be cynical, but I do feel like there are some recordings that seem as though the goal is to mislead, where clearly artificially enhanced/modified demos (without being disclosed and important context omitted) are presented as if that’s just how the board sounds in person; personally I don’t really agree with doing things that way, but I guess those that fall into this group aren’t going to read something like this anyways so maybe not worth dwelling on lol. I also don’t want to be the sort of person who tries to jump on others for doing it and such, I just wish it wasn’t really a problem. I don’t assume most are like this, but I have seen some pretty popular examples sent to me that fall under this category and I’m not going to act like it doesn’t bother me lol.
Most times when I’ve heard recordings that make me skeptical or feel unrealistic, I don’t believe it’s intentional; rather a lack of understanding, lack of proper equipment, or poor technique, a lot of the time all of the above. In my opinion, recording a board isn’t as simple as slapping a mic above something and calling it a day, I think it’s a somewhat multifaceted subject that demands a bit more time, energy, and trial and error than most really give it. That being said, I’m also incredibly picky about my hobbies, so of course I’d feel it demands more attention lol. Regardless though I think it’s still worth talking about (even if maybe I care more about it than I should), and to start off that discussion, I think it’s important to make clear some of the limitations that you’ll encounter first.
Inherent Limitations:
Most obviously, it’s a recording, and not anything that can actually mirror or replace experience sitting in front of a keyboard and using it. Everyone can understand this (I hope), but sometimes I look at comments and wonder lol, this isn’t to say you can’t aim to make a recording as close as you can feasibly get though. I understand why some completely dismiss recording boards as a fruitless effort and not anything that can offer any valuable or useful insight/information, but I personally disagree. I do understand there’s a limit to how actually realistic and useful information you can get out of a recording, but just because it can’t be a perfectly accurate representation doesn’t mean what can be done should be entirely dismissed, and this is coming from someone who’s main priority for a keyboard isn’t even sound. I think it’s unwise to fully trust in sound demos no matter how potentially good they are, but I think it’s also equally unwise to just write off every demo because the flaws exist in theory.
As an example, in the case of the demos I upload on my channel: you’re only getting the sound of my current build (which is built from my experiences with other boards in my environment to my preferences) in it’s current condition (which is influenced by it’s current environment), on my deskmat and my desk in my room/space (which makes a big impact, including type/material/size/placement/things on desk/things and materials in room/floor/etc), with my equipment (mic/mount/preamp/interface/cabling which can be wildly different from others and can have variance between units of the same model and include aspects beyond what most consider for recording) in my signal chain (with it’s own synergy) placed and configured to my preference (mic placement and mounting, polar pattern and preamp configuration, line gain, digital settings, recording and editing methods, etc) determined by my own hearing and listening ability (hearing is innate, listening is a developed skill based on experience), then processed and encoded by YouTube with their lossy codecs and other modifications (which they keep adding garbage that fucks with your audio), played back to you through your device with any of it’s additional processing (lossy compression in the case of Bluetooth, sample rate conversion, EQ or loudness compensation, etc), delivered though your listening equipment (which is different for everyone in individual equipment along with overall chain and synergy and has it’s own coloration and ways it modifies sound) which is also influenced by your environment in some cases, into your ears (which are somewhat different for everyone, also impacted by your current conditions) and processed by your brain (again, different for everyone) against your own expectations/experiences/biases/listening ability (you get the idea at this point).
There’s a lot of points for problems to manifest, and why it’s so hard to get a truly good recording and listening experience from start to finish, there’s a lot working against you. There’s also high potential for varying degrees of quality loss at each step that can mar the final result, along with making true apples to apples comparison extremely difficult (you can try to set as many variables equal or similar as you can, but there’s still only so much you can do). It’s also a good example of how much is dictated by preference and individual setup. All this being said, there’s still plenty of ways you can work to reduce and somewhat mitigate these inherent downsides, and when done properly, you can still end up with some convincing results. I think it might be helpful to break things down into some categories.
Recording Equipment:
I really had to think if I was going to talk about equipment or technique first, and while I think they’re equally important, in order to maximize technique, you need to carefully select and properly understand your equipment. I will say right off the bat that more expensive or boutique does not always equal a better result, I do generally think you get what you pay for, but you need to make sure you plan your setup for the right use case and matching and pay for the right equipment that meets your actual needs; use it in the right way that actually allows it to perform for the most optimal results, which is easier said than done lol. Most all gear is not a one-size-fits-all-jack-of-all, there are many different routes you can take to achieve a satisfactory setup for your needs, but there are some aspects I feel are generally better than others when it comes to recording keyboards, but someone skilled in recording can make most setups work to be honest. It’s easy to get overly fatigued from the amount of choice and potential options and how everything will fit into your setup and mesh together, so I think some narrowing is required to keep some level of sanity lol. I will not be explaining every small aspect that goes on with everything if it’s something easily searchable or irrelevant to recording a keyboard to reduce bloat, Google can be your friend, but they’re probably a friend you should take with a large grain of salt since there’s often a lot of nuance missing from general audio discussion. Let’s start with the most critical part of a recording chain first.
– Microphones & Mounts
There are a lot of microphones out there. I mean a lot. From my own experiences, I personally feel is the best type for recording a keyboard is the Large Diaphragm Condenser (LDC) microphone group for a variety of reasons. This may not agree with some who might argue the Small Diaphragm Condenser (SDC) is more suitable for this application, which are my runner up option depending on setup, but let me explain why I prefer LDC more first. In theory, SDC would be best suited for a keyboard given their more even response, lower proximity effect (because it’s generally more rolled off in the low end) and reduced bass distortion, more uniform polar pattern, and faster speed/tighter transients over a LDC, they’re largely used as instrument mics at distance in the field which more closely aligns with what a keyboard would fall into compared to the more typically upfront uses for a LDC (but LDC are also used for instruments often as well). However, in practice, think LDC actually sound more accurate to ear with a keyboard. LDC’s generally have lower self noise which can be helpful for recording a quiet board at distance, more flexibility with polar patterns for more forgiving positioning, a lower resonant peak than SDC which sometimes works better for a keyboard, and I feel that their transient response is more organic with a keyboard compared to SDC, I also believe the higher frequency rolloff on a LDC works in it’s favor for a keyboard as that’s closer to how we would hear it in person. With the recording environment being the average computer desk, an LDC offers better capture of the space (which is important to getting a convincing recording) than a SDC does (although a well placed SDC may offer more rejection of undesirable reflections than a LDC) from my experience. With the right model and setup, I do think SDC can be equally viable, I just personally prefer LDC, and that’s based on my own listening and recording experience with keyboards more than anything else.
***NOTICE*** Everything else written in this article will be under the assumption/expectation that recording will be done with a LDC in Hyper/cardioid pattern. Maybe at a later date I will add recording notes specific to SDC and some dynamics, but for now it will be LDC focused I have the most experience with it. It’s not like a lot of this can’t be applied to other types of microphones but it may not be perfectly aligned/accurate/relevant, just keep that in mind.
Generally, I wouldn’t suggest a dynamic mic for recording a keyboard. Most just aren’t up to the standards of what I’d personally consider ideal for this application, even if there are a few outliers that can work well. I just think that the mass majority of dynamic mics out there do not offer a convincing capture of a keyboard and add too much sonic coloration in various ways to the recording. Some aspects of coloration come from shifting tonality and their off axis performance, their less sensitive nature compared to a condenser, and their inherent compression in sound as well. Personally the only dynamic microphones that I have tried (and I have tried most of the common and some uncommon options over the years) that I would feel are suitable for recording a keyboard at the same level as a moderately good LDC would be a Sennheiser MD441 or Beyerdynamic M201, and even then, I’d rather have a LDC or SDC. Another thing that’s somewhat neglected in terms of planning a setup with a dynamic is the preamp, generally the performance of a dynamic hinges on a good preamp, and in most cases the quality of ones integrated into an interface are just insufficient where there’s a bit more forgiveness with a LDC or SDC for a mediocre preamp. I just dislike seeing the various Shure or Rode or more broadcast/stage/vocal focused dynamic microphones be used for keyboards because they just impart such an obvious and inaccurate coloration to boards and are often underserved in terms of preamp and placement as well. Just not the sort of microphone you want for this application. On a similar vein (and sort of a subcategory of dynamics), I would also consider avoiding ribbon mics. They have some of the similar characteristics of dynamics taken to a larger extreme in coloration and preamp pickiness, I do think they generally offer a higher level of technical performance though in terms of detail and nuance which does make me slightly prefer them to dynamics for keyboards even if neither are ideal. But hey, wait a minute, didn’t you use a ribbon in your recording setup? What gives? Yes, but only as an extreme difference example, something that was only there for contrast, clearly stated it wasn’t accurate compared to the other mic used, and just for fun lol.
One last thing I’d like to touch on are shotguns, I think in most cases they offer a misleading result for a keyboard, rejecting a lot of off axis noise and only capturing a small direct point of a board’s sound. This can be interesting and maybe desirable in some case as it can unnaturally reject overall board resonance or undesirable room reflections and noise and instead act as an audio microscope directly on the sound of switch interactions, but this is just pretty far from how people actually hear a keyboard during use. You can try and mitigate this by really giving your shotgun some distance to capture more of the overall board, but you’ll still be wrestling with some of the same limitations/coloration from it, so I’d just avoid going for one unless you really want that isolated microscope sort of sound as an alternate option or just for fun/research I guess. I would also suggest avoiding the XY stereo capsules, and stereo in general with AB pairs, but I’ll comment on that later.
For my own setup, I elect to use 2 different microphones as an attempt to give more information and insight into a board’s sound, because it allows me to pick 2 microphones with different characteristics and strengths/weaknesses, along with having multiple options for placement as well (which may be even more useful if used with 2 of the same microphone to showcase the difference in sound from different positioning like sitting position vs directly overhead board, or sitting position vs distance/overall room). This does drive up cost a lot depending on how you do it in terms of another mic, and the cost of adding another channel (from your interface + preamp if applicable), but I think it can be valuable if done right. I considered doing 3 mics + camera mic, but aside from more added cost, you may run into issues with more than 2 mics nearby as you might end up with them disrupting the sound of the recordings as you try to fit 3 mics without blocking eachother at your desk which could cause strange reflections or side effects in the sound from their physical presence. If you did this with 2 separate interfaces this may also end up increasing difference in terms of trying to sync up interfaces with recording and potential sound differences between the interfaces. Will get into interfaces and preamps later though.
All this rambling and I haven’t even really talked about what characteristics to look for an a LDC for keyboard recording yet? That’s how it is sometimes lol. Personally I think you’ll want to consider either a cardioid or hyper/supercardioid for recording boards, from my own testing I feel as though that’s the best balance in polar pattern for accurate sound for the techniques I use, but there are occasions where a figure of 8 pattern microphone could have some benefit if utilized correctly. In terms of overall mic design, I find those with more compact size and lower weight to give more flexibility in fine tuning your recording sweetspot, allowing you to dial in your settings at a desk. With hyper/supercardioid it can be helpful for dealing with a noisy room, but you might find you have to back off a bit to reduce proximity effect and capture the full board vs cardioid (as with cardioid at the distances you should be miking at it’s not a big issue). I do feel that it’s important to capture some of the room and space along with the board, so having a balance between focusing on the board but not attempting to fully isolate the pickup is something I consider in what to choose. I also find that a mic with a good focus on a nice head basket/internal mesh design will better reduce some internal reflections within the mic itself and allow you more flexibility in angle as well, but this is hard to tell without actually testing a lot of different mics. I could talk about various aspects like frequency response and sensitivity and whatnot, but to be honest the most important thing is just how it sounds in person and the only way to know is to try, and I think if you can find a microphone that’s relatively flat ish on frequency response and not overly difficult to drive from a sensitivity perspective you should be fine, it’s more important how it actually sounds to you in your setup. The only way to truly know is to find out yourself, so if you’re really picky about things it’s best to order or demo a few options and see what’s most convincing to you. I wouldn’t pay much attention to most marketing blurbs about capsule size and material (although a good capsule is extremely important, a lot of it is just marketing and you can only tell a good capsule from the sound really) and any sort of internal dsp or compensation (in fact I’d avoid any mic that advertises itself as a digital or DSP mic), and so on. There’s a lot of aspects that can help to make a good mic great, but really when buying a finished product you just can only really tell by hearing it first hand. One important thing I will note actually is the presence of a transformer in the mic, those generally cause a mic to sound different and generally more colored due to the transformer itself, but also the difference in the load and drive presented to the preamp/interface it’s connected to. Some microphones can sound fairly neutral and realistic with a transformer, others may only have it for added coloration which isn’t desirable in our case. I do think that if your goal is overall most “transparency” no transformer will generally sound more immediate and raw vs having one in place, but that’s not always a good thing. I’d generally say unless you really know what you’re doing to stick with transformeless mics as they often offer better bang for the buck and generally are more neutral (some may consider them more lean though) than their transformer coupled counterparts. Really a lot of quality LDC can do the job here, there is not one magic pick, you just want something overall balanced and high quality in most aspects.
Since I know some may want to know my suggestions on what microphones to consider, here’s a few that I think are good starting points at a more budget price point (am forcing myself to limit it to only a few per category lol). NOTICE: If I denote Used in front of something, that just means I would suggest only buying used, and not new as I didn’t think it was good value new. I would highly suggest buying used when you can to save money with higher end gear specifically if you can. This is for all times I use it for the rest of the categories as well. Anyways:
* CAD E100SX – what a sleeper, such a good mic for the price, the E100S was already a sleeper and now the SX is at a lower price while still being pretty solid (big discounts on them now), hard to beat for this application being compact, reasonably neutral, and surprisingly technical for the money, great option to start and stick with to be honest.
* Rode NT1 – for the price you can get them, very good on resolving detail, they are a bit colored when it comes to how they portray dynamics and detail (as well as a bit sharp) but I think they’re just a good value and are well suited for a keyboard.
* Lewitt LCT 240 – A bit brighter more presence boosted but overall more on the flatter side of things, a bit less technically capable sounding to me for capturing space vs the aforementioned though (but more neutral).
For the step up from that, I would say these would be 3 picks that I think would be a worthwhile upgrade from the above and realistically the highest that most should consider going with even if they’re recording keyboards often:
* Roswell Mini K87 – Extremely well tuned for the money and very thoughtfully put together, a bit more immediately flattering than the next 2 mentions but in a way that still feels organic and faithful to me, pretty technically capable as well, just great, though if you want the most neutral option in the range there might be better options, I just think this is a really good balance and pretty workable.
*Austrain Audio OC-16 – Neutral and no nonsense, some might immediately find it a bit dull due to it’s unhyped nature but it’s got good nuance and is smooth without missing out on technicalities, it grows on you overall and I think it records boards pretty honestly, also slightly warmer smoother which is nice, maybe a bit dimensionally flat though.
* Used Audio Technica AT4050 (or new 4040) – A good reliable workhorse option that really falls into the jack of all master of none and owns it, it’s just pretty even and balanced with not much to write home about, but that’s part of the value in it as well, the results seldom turn out poorly. 4050 are common used for cheap so the value is high, I’m less a fan of the 4040 though personally.
I selected these options based on my personal experience with them and their sound, but also considered value, consistency/QC, pickiness (in environment, positioning, and source gear), and also accessibility and reproducibility. I could’ve gone for some pics that might be more niche by smaller mfg or one man bands that offer great value and performance but have a lot more variance and limited availability, or more options that were easily available but didn’t quite meet my standards in sound, I think these options are a good balance for this application that stay right within the realm of reasonable in most respects.
If you’re considering going higher than that 500 ish range, you should likely have a better idea yourself of what you’re after (likely something more specialized), and better have all other aspects of your setup up to snuff first, there’s also so many awesome options around that 1k+ mark that it’s hard to narrow things down, I will give my picks for that at the end of the Equipment section since it involves the preamp and interface more than the rest. It’s also beyond the realm of value for most people, they aren’t really going to appreciate some of the benefits unless they’re really into this hobby I feel. If you really want more info on this realm leave a comment or PM me I guess lol. For my own setup, I’ve gone through a lot of mics, more than I should have over the years, and I guess if I had to list some of the most impressive mics for keyboards that I’ve tried (either directly on keyboards themselves, or on enough applications that I know how they respond/react), they would probably be as follows in order of preference:
* Josephson C716 – most realistic sounding mic I’ve ever had, to the point where it’s a hassle to deal with in every other respect
* Neumann M49/149 – close to the above but more immediately pleasing and has it’s own character.
* Sanken CU44X MKII – extremely clean and neutral that makes things pop, very dimensional and slightly ethereal, low off axis color
* Pearl ELM-C/C22 – very flat and even up until some treble rolloff while still being convincing, slightly relaxed, just well balanced in most categories and because of it’s capsule depending on positioning you can use it to better isolate something and reduce reflections on one axis specifically.
* Gefell MT 71 – a modern take on a “vintage” neutral mic that’s a great all rounder
* Probs Schoeps/DPA SDC – what you think of when you think of neutral, maybe not truly natural though, but dead flat and garbage in garbage out.
MORE ON THE HIGH END LATER, THIS NEEDS CONTEXT. These are way overkill for recording keyboards, it doesn’t stop me because I’m way too picky myself, but these make little to no sense (negative sense?) so keep that in mind lol.
For any of these mics, you will want a good mount, preferably one that is isolated/detached from the desk you type on, or if it has to be connected to your desk one that has internal springs and is solid (includes some thought toward reducing overall vibration). A crappy mic arm will transfer more desk noise into the shockmount and mic which just sullies your recordings sometimes, and also may ring or make noise of it’s own during recording. On the note of a shockmount, you want a good one for the same reason, sometimes the one that comes with the mic (if it includes one at least) can be good enough to just stick with, but it’s worth looking into better options if not. Specific suggestions will depend on the mic. I do think that when it comes to arms and stands, the more flexibility in placement the better. Depending on your flooring material, you may even want to work to isolate a floorstand from the floor as well, but all depends, worth considering to improve sound later on.
– Preamps, Cabling, and Conditioning (Optional)
Truthfully, unless you’re venturing into the higher end or upper midrange market, or want to give your current setup a bit of a boost, you won’t need an external preamp. I wouldn’t even consider one with the mics I’ve explicitly suggested in the step up category, only the beyond category which I didn’t even give specific picks for. I do feel as though most preamps built into modern interfaces are what I’d consider “adequate”, but as you increase the quality of microphone you’ll find that not having a good preamp can become a significant bottleneck. For myself, I’ve gone though so many preamps that I really shouldn’t have, some were absolutely worthwhile and elevated my setup, others just left me questioning why I was even bothering, it can be quite a mixed bag, and something that is very setup dependent. A tough pill to swallow is that most cheap preamps just aren’t worth buying outside of if you’re looking for a slight amount of coloration or absolutely need more gain (which we aren’t/won’t here in most cases), they just aren’t that much better than quality interface preamps enough to justify themselves for this application. I personally feel that the actually difference making preamps tend to start more around that ~1k ish mark, and you’re already balking at that figure then you will want to move onto the next section lol. I’ll be brief on this section so it’s easier to scroll past.
If you’re still here, then my suggestions for a neutral (but not clinical) leaning preamp that will make a noticeable difference without venturing into the hard to find/boutique and/or exorbitant price range (and also commonly available used since you should buy used when you can):
* Used Buzz Audio MA2.2 – A really nice balance of being largely neutral without sounding sterile, it has a hint of bloom and weight to it’s sound but is also very technical, not being balanced makes it a bit of a risk in terms of picking up noise but if you can make it work, it’s very nice for this application as a solid reference.
* Used A Designs Pacifica – It’s got a bit of a more vintage vibe but surprisingly not enough for me to consider it all that colored, honest yet having some noticeable character of it’s own that will mesh well with most of the aforementioned mics by adding a bit more body and depth while elevating technical performance, eversoslightly softened at the top, maybe a bit more grand.
* John Hardy M1 – Clean but not dead, has a bit of character in terms of being a bit more forward and aggressive but in a pleasing way, just very reliable and very capable in terms of technical ability for the money, barebones but worthwhile and super versatile.
* RND 5211/5012/5017 (or some sort of Portico) – Neutral with a bit of bloom is how I’d describe modern Portico preamps, they’ve got a slightly softer smoother touch but still doesn’t feel overly colored, and generally works well for everything, if you’ve got a dryer setup this is a very nice option to consider, the 5012 is the best value if you can find one, 5017 and 5211 also good.
* AEA TRP3 – While it might be intended for ribbons, it’s actually pretty great with most mics I’ve thrown at it with loads of gain and generally neutral, pretty punchy and has a lot of control, although can sound overdamped/dull with some pairings, make sure to do some research for pairing.
* Used Grace Design m101/201 or DAV BG1 – A bit more on the brighter and slightly thinner side of things, these can be great value on the used market moreso than the aforementioned and are also still versatile, although if you feel as though your setup already leans a bit bright or leaner then I don’t think these would be the pick to try and rectify that, but good still.
I do think that for max value getting a 500 series preamp with a lunchbox or some sort of rack conversion would offer the best value, but truthfully I don’t really have much experience with 500 series modules since I mostly ignored them in my time doing professional work, so can’t comment, but likely worth exploring as used they tend to offer more bang for buck if you have an enclosure for them. There are also lots of other cool and worthwhile options out there, but are not easily available (or too expensive for this application), if you could find a reasonably priced Crane Song, Gordon, Pueblo, Forssell, NPNG, etc then those are amazing options for this application, it’s just not really worth discussing since I think most people, even those who would find it worthwhile to invest in a preamp, would likely not find the justification to buy lol. I think in general for this application you want something with a moderate amount of gain (at least above 60db), something that’s focus is just quality and simplicity rather than a boatload of features at the cost of quality, and something less focused on character and coloration (so maybe skip your vintage Neve/API/etc stuff). I would skip things with a built in eq or compressor something like that as even when not in use it can still be passing through and adding coloration (and adding extra unused cost as well). When it comes to transformer units (for both input and output stage), I don’t think it’s as clear cut as microphones personally but I do still think that no transformer is generally going to sound more raw and technical than something with a transformer but more mixed here as some of my picks above do have transformers and some don’t as I think it really depends. Generally skip tube based stuff, there are some very neutral tube preamps out there but they’re harder to come by and typically very high end, in most cases it’s going to be more understandable coloration and hassle. It’s hard to really make suggestions without knowing the mic and the interface, you have to think of everything as a system, in that it has to fit in and synergize with your system to lead you to the final result you want. It’s also something that you really do need to have a high quality mic and interface to take advantage, especially an interface with true direct line inputs, if the signal is still going through the internal preamp of the interface, it’s not worth buying a preamp until you get an interface with direct ADC inputs. A proper preamp requires more thought than most people would likely want to put into their system lol, don’t consider these unless you’re at a higher level than most with microphones and interfaces. I do think that if you did want to use with some of the aforementioned mics, a Grace or AEA preamp used could possibly not be a bad idea if you can get a solid used deal on one.
I’ll take this brief opportunity while I don’t have the attention of most to talk about cabling and mention how that can and does influence the sound of the final result, crappy cabling just tends to sound dead and flat in comparison to even good studio standard wire, and also has a higher chance of picking up electrical noise and interference which can be the bane of your existence when recording. If you want to care for XLR, I’d pick up some Mogami Gold if you want to lean your system a bit warmer and thicker, or Canare Twisted for something more down the neutral range. I don’t feel like commenting further, or on power or USB cables since I don’t want to invite myself on that, it does matter, but it is marginal for this application, just get some good pairs and don’t think about it again and you won’t have to worry.
Power conditioning/regeneration can be very worthwhile depending on the quality coming into your house and other things you have plugged into your circuit. Cheap power strips or conditioners can actually hurt quality, getting a good audio focused conditioner (NOT UPS) and/or isolation transformers can reduce noise (both lowering electrical noise floor and mechanical hum/vibration from transformers or capacitors) and improve subjective quality, but it’s typically still a matter of give and take, I notice in the past that I would get improved clarity, more resolving power, more depth, but might take a step back in slam, weight, and width, it all depended on the conditioner/regeneration. I used PS Audio, Shunyata, Audience, and Equi=Tech and had good results with different equipment with all, the bog standard stuff didn’t do much though for me most of the time (and often was worse than directly into the wall from a sound perspective). I don’t have any power filtering/conditioning/regeneration for my current setup, but it absolutely needs it, so it will be happening in the future lol. I can notice small shifts in sound quality and noise during specific times of the day, just a good indicator of crappy power quality lol. One easy and free (ish) thing to do at home is to isolate your audio equipment (if it’s powered from the wall) to one circuit with nothing else on it, and your computer and other electronics to another, you may notice an improvement in quality and reduction in noise that way, but not everyone has the ability to do that easily (layout of outlets in house, or maybe another noise source polluting a circuit you were going to use).
– Interfaces/ADC
Personally, I place great importance on source gear for getting a good final result, and while the difference between preamps and interfaces/ADC tend to be less than the difference between microphones and technique at first glance, underneath those small differences make a big difference to the final result. An interface is typically the thing that’s both amplifying your microphone and converting the signal to digital, and getting an actually good result with that is not as easy as some people think it does. There’s a lot of potential loss that can happen here when the interface/adc and preamp become a bottleneck in higher end setups, so I think it’s worthwhile to be mindful of that. Most often with lackluster preamplification and conversion you end up with flat and lifeless results that lose out on dimensionality (even in a mono signal!), texture and low level information, and lackluster dynamics, all of which I do think are important for a convincing keyboard demo (due to properly capturing the keyboard and the space it’s in). If you just skipped the preamp section, fair enough, but there’s still some information still worth reading in there even if you’re not going to get a standalone preamp. You want to properly match your interface/adc with the rest of your equipment, and getting good system synergy will lead to an overall better result, use your equipment to bolster their combined strengths and make up for their weaknesses.
I do think that interfaces suffer from the same problem preamps do, there’s good ones for cheaper, and really good expensive ones, but the middle range just aren’t worth it outside of niche situations imo. In terms of what to actually consider when getting an interface, I would say the most important aspects to me are overall sound quality over everything else which is combined preamp and ADC performance, after that would be usable gain range that doesn’t limit how you record, and true/direct line inputs for an external preamp (though these are rare on cheaper interfaces, most of them can do line in but it’s still going through a preamp circuit which isn’t as good). I think going with a 2 channel interface yields the best balance of future flexibility without going overboard, but if you only ever plan to use one mic, then a single channel is fine. I’d personally shy away from going with over 2 channels since you are getting less SQ per dollar that way unless you actually plan to use that I guess lol. Also less on the importance scale would be DSP features since we won’t really be using those or other sorts of methods for coloration (like UA for example, DSP can be helpful but it hurts the value in this case and ignoring that their interfaces aren’t great SQ per dollar because of that), or killer USB drivers over everything else and/or excessive digital IO since the latency and such really doesn’t matter in this case either (I feel this way about RME interfaces, fantastic USB drivers, but sound quality is more mediocre). My picks for a reasonably good option for most use cases (but also tailored to suit most of the mics mentioned previously) would likely be:
* Solid State Logic SSL 2 MKII – A more clean yet a bit sweet and forward sort of sound, when the pairing is right this is really good interface for getting nuance at the price point without sounding offensive, good depth and great dynamics even if it’s a bit thinner in the low end sometimes, great USB drivers.
* Audient ID14 MKII – A good balance for the price point moreso than some other options, fairly neutral sounding with a slight bit of warmth and more relaxed treble, nice resolving power as well, it’s just presented more organically than some others on here even if it doesn’t have immediate wow factor as the SSL, USB drivers great.
* Black Lion Audio Revolution 2×2 – A bit more character than others on this list but not to the point of disrupting anything, more forward and pushed in the midrange, a bit more weight there as well, also more energy than some others, although a bit noisy and limp when driven harder, USB drivers okay, it’s a bit more unbalanced but still worth a look with the right mic.
* Motu M2 – Neutral, clean, even sounding, somewhat more dull than the aforementioned options but not as dead sounding as something like a cheaper Focusrite, preamps good enough to not worry, a safe bet with most things although maybe pair with a bit warmer mics.
There’s actually a limited amount of higher end audio interfaces that I thought were worth it in terms of sound, most of the time the best conversion are with dedicated ADC which don’t include a preamp and for most that’s just going to add too much cost and space. It seems like somewhat of a neglected market, but I also understand why (if you’re really putting this much care into your recordings, you’ll be using outboard equipment and want a standalone ADC). I think these picks are the few that I’ve tried that I thought had a good balance between internal preamp and ADC performance while still offering good value per dollar:
* Used Prism Lyra 1/2 – By far the best SQ interface I’ve personally had, it’s it’s fairly neutral to me, a slight bit wider, smoother, and almost a bit contrasty when it comes to presentation of depth and dynamics, ending up natural adding a slight sweetness/niceness to everything, mediocre USB drivers though.
* Used Lavry AD11 – Older but really really good, this is a bit more colored with more character being a bit warmer and more relaxed but pretty organic and not something I’d consider too colored for this application, very nice weight and depth even if it’s not the most detail forward, USB is old though and may give a bit of headache.
* Sound Devices USBPre2 – A higher quality swiss army knife that’s very neutral, very even, very technical, but without sounding bland and flat like some others, it just really works with mostly anything and is consistent, solid USB.
* Neve 88M – This interface does have more character than others on this list, but it’s on the edge for me, and while the ADC aspect of it wasn’t incredible to me, the built in preamps of the unit end up getting it a place here as they’re awesome and what you’d expect for Neve. I do think that maybe a bit too colored for some pairings though, so you’ll want to keep that in mind, USB alright.
After this you really just want to consider a dedicated quality ADC, some brands that I’ve tried and like are as follows, won’t really elaborate further because this isn’t worth it to most: Crane Song, Lavry, Forssell, Merging, Prism, Weiss, EMM. Big money, big hassle, but big sound and big worth to someone ultra dedicated, skip for most people though lol.
The higher the quality of the ADC/Interface, the more picky it will be about everything else, and it may be worth investing in something like a USB isolator/filter/regenerator if you run into ground loops, but also to squeeze a bit of extra sonic performance from a nicer ADC/interface, personally some reasonably priced options I’ve tried are the Altor HISO USB (TI version, paired with a powersupply like an iFi iPowerX) or a JackAudio ISOUSB211 (with a similar powersupply, the new TI chipset is really good for the price), or more like a regen product like an Uptone ISO Regen. For a step over that, consider bypassing USB altogether on your interface and using a quality DDC from someone like Mutec can be a pretty noticeable improvement. I will refrain from mentioning USB cable quality (who knows, could be worth exploring), just use a decent one at least lol. All of this is only something to explore if you’re closer to the upper tiers of performance though, it’s not going to be worthwhile on an entry to mid level interface from my experience.
– Assembling the Chain:
Now that you might have a bit of an idea on some of the more individual pieces that go into a quality recording setup, but now they need to be put together correctly. Well what is the correct way to pair them? That’s going to be up to your ears, and your environment. Really a lot of this stuff is trial and error, maybe consider buying things with the ability to return if you don’t like the result, or if you have the funds consider trying multiple things at a time (I would focus on mic first for this) and try them in your own space and conditions with your own opinions. Sometimes things that seem like they’d pair well just don’t pan out and lead to meh, sometimes the questionable pairing actually works out and leads to an interesting and worthwhile result. All just comes down to real world experience, your experience. Maybe a bit of a copout answer, but really that’s just been my experience lol. All the suggestions that I’ve mentioned will all mostly work well together (some better than others) based on my own experience, but there’s no real set rules for pairing to be honest, you just need to go by past experience if you can, and if you don’t have any, you should get some now by experimenting yourself.
I had mentioned goals, so out of what I mentioned, what would I pick for each goal within reason?
* Showcasing/documenting current build: E100SX / Mini K87 -> ID14MKII
(I think the CAD and Roswell are natural and capture space well, not the most neutral but do a good job of faithfully capturing the build, Audient evens them out and a solid base that pairs well with both)
* Creating entertainment: NT1 Signature / Mini K87 -> Revolution 2×2 / SSL 2 MKII
(These combos will really pop and slightly embellish the sound in a pleasing way without being too disingenuous, just pleasing and more forgiving in placement as well)
* Reviewing a product: Lewitt LCT 240 / Audio Technica AT4050 -> Motu M2 / Audient ID14 MKII
(More on the neutral and flat for honesty and replicability, along with lower variance, these combos will lean a bit bright but they’re more general workhorse that will work well with a wide range of use cases)
* Comparing/contrasting: CAD E100SX / Austrain Audio OC-16 -> SSL 2 MKII
(Getting a bit more of an even tone while capturing a lot of detail is important for picking out smaller differences, this chain does that well giving you a better picture of the differences)
* Just cause: why are you still here lol
What about the higher end preamps and interfaces? Yeah, I realized that after writing too, I included them but not anything about mics that actually would take advantage of them that were attainable, I should really mention some here so it’s not useless information
In terms of higher end mics, I will separate some picks that I think you could get away with only using a higher end interface and not be leaving too much on the table (along with being largely available and consistent), and ones that I think require that next level of attention and care. Let’s start with the interface only needed picks:
* Gefell MT 71S / M930 -> Lyra 1/2 OR Neve 88M
(The Gefell is in my mind an improved option over a TLM103 which is what I’d have suggested otherwise, to me it sounds more even and works better with more distant sources than the 103 along with being very compact, it’s very technically capable with lots of resolution and depth, clean but not limp or dry, fairly natural but not as exciting as the 103. Pretty high value and one of my favorite options before getting into top tier mics. I prefer the 71S a bit more as the 930 sounds a bit more thin but both are good and similar. Pairing with the Lyra will focus more on the clarity and effortlessness of the mic, whereas pairing with the Neve will offer more weight and density which can help in some setups)
* Sanken CU-55 -> Lyra 1/2 OR USBPre2
(Sanken mics will focus on getting the highest level of rawness from the source as they can, very very flat sounding and very critical of mistakes, but worth it if you know how to record properly and have tuned the rest of your setup. I don’t think this is as extreme as the CU44 but it’s very good for the price, lots of detail and low level nuance that others smooth over in this price point. I would say the Lyra is a solid pick for capturing the nuance while slightly sweetening and relaxing the sound, the USBPre will be somewhat less technical than the Lyra here but in exchange it will offer a more even and upfront sound that will be great if you’re after most neutral pickup)
* Used AKG C414 XLS -> AD11 OR Neve 88M
(A great workhorse mic that generally goes well with everything, although might not shine with most, I wonder myself if I should include this as it’s never truly blown me away, but it’s also never failed me either, and considering how cheap they can get used, I think it’s worth a mention, it’s fairly even slightly on the more tame and dry side of things, but they’re also fairly forgiving compared to the aforementioned options, so I think that in itself is valuable. I like to pair them with a bit warmer source to add a bit more sweetness and depth to them without emphasizing dryness and I think the Lavry or Neve do that well)
Getting into the “you really shouldn’t be doing this unless this is your job or recording is a main hobby” range, here is what I’d personally pick for readily available mics (mostly). I so want to swap out some of the preamps and ADC here as you could do better for these microphones, but I am forcing myself to limit to actually available and not too far gone levels of suggestions lol:
* Josephson C705 -> MA-2.2B -> Lyra 2
(I hesitate to mention it here because this is such a demanding microphone in terms of both preamp and ADC, but I think this setup could make it work for the most part. This is an incredible microphone, not too far off from the C716, it’s extremely accurate and technical while still being largely convincing and mostly neutral, it will pick up everything and all the problems in your space like the 716 will, so it’s only for the most meticulous, but man is it awesome. Only downside is that proximity effect can be high because of it’s single sided diaphragm. The Buzz is a great pairing overall, complementing the nature of the mic and letting it’s resolving power shine while adding a hint of character and not sound dry. Lyra 2 also has a similar impact of letting the nuance of this mic through while imparting a bit of wetness)
* Used Wunder CM49 JFET / CM7 FET -> Pacifica -> Lyra 2
(Natural, sweet, and slightly forgiving yet almost as technical as the 705 makes this an awesome but kinda expensive pick, it’s convincing but has it’s own flavor that’s fairly unique from other options, hard to describe but it just sounds polished out of the gate but not processed if that makes sense. If you can get the CM49 that’s the one to have, though I felt the CM7 was also good but just not as captivating. The Pacifica also feeds into these mics strengths well, it does end up leaning the overall combo maybe a bit more sweet and rich than preferred however it just sounds so nice that I think it makes sense lol. Lyra just feeds into this as well, though again, maybe not the most honest thing. It just sounds too good alright, I’m sorry lol)
* Sanken CU44X MKII -> MA-2.2B OR JH M1 -> Lyra 2 OR AD11
(Extremely lively and full of texture and depth, very flat with really low off axis coloration makes this capture everything in a critical but not clinical way, it’s not something that’s as organic as the Josephson but it might be more by the books neutral, and that has it’s own value. The Hardy follows a similar goal, being very neutral with little traditional coloration, and it’s also fairly fast and punchy which works well to mesh with the strengths of this mic. Lyra 2 does add a bit of coloration but only slightly and it helps to tame some of the extremes, the AD11 would be the option if you want to slightly relax and add more body while still capturing most of what make it unique, so more of a compromise but a worthwhile one in some applications)
* Pearl ELM-C -> MA-2.2B OR TRP3/M101 -> Lyra 2 OR USBPre2
(This mic feels like it takes the strengths of a ribbon and a condenser and puts them into one, leading to a very organic and nuanced sound that’s more smooth and relaxed but still neutral, it’s really engaging although can be somewhat dull on the wrong source. If you position this well, it can really cut out nasty reflections on one axis. I think the Buzz wakes it up a bit but also keeps it’s organic and evenness while adding a bit more energy and bite especially in the treble, the AEA or Grace in comparison are less nuanced but offer a similar sort of experience with the AEA being very neutral and fast overall, the Grace leaning a bit brighter and cleaner. Lyra 2 helps to preserve some of the organicness and nuance more than the USBPre but that keeps a bit more of the neutrality of this mic)
* Used Neumann U89i -> RND 5211 OR Pacifica OR BG1 -> Lyra 2 OR AD11
(Aside from the M149 this is likely the best sounding pick for keyboards from Neumann, it’s more even and natrual sounding than the 87 while still having some of the characteristics that make the 87 special, a bit slightly contrasty and “finished” sound but it doesn’t sound as sculpted/preprocessed, and is great for presenting a lot of information in a smoother more casual sound, it just sounds great with most things and hard to argue with that. I think the Neve or Pacifica adds a bit more character in different ways with the Neve offering a bit more neutral smoothness whereas the Pacifica is more crafted and deep sounding, Lyra 2 for reasons mentioned above but AD11 to drive it more toward what a vintage 87 can be like if you’re into that)
You see mostly Lyra 2 here as a suggestion, and that’s just because I think used that’s the best value per dollar in sound ADC you can get easily (even new it’s reasonably priced when on sale), you will easily 2-5x price for that next step up. Though a Crane Song Interstellar Quantum is within the realm of reasonable for this equipment and I would much prefer that over the Lyra so keep that in mind if you’re looking at a new Lyra and planning on not using it’s built in preamps. Other options aren’t as accessible that I actually like.
So how did I end up with my setup? I talked about it a bit in my schizo rambling on my few recording setup videos, but I mostly was just drawing from my past experience doing commercial mastering and hobbyist recording, I’ve gone through a lot of different gear (Gear Acquisition Syndrome hit me hard when I was younger and I wanted to continually advance my craft lol) and I’ve ended up having a lot of hands on experience within the audio realm in general. If you’re interested in more on the listening side of things, check out my other articles on here and the forum as well, but I’m just using my forum as a place to publish this, how selfish lol. I’ll skip some of the backstory esque experiences I’ve had with recording and fast forward to when I started this channel using an AKG C414 EB into an API 512 into a Lyra 2 (and a Beyer M160), and at first that chain performed well enough for me to not care for the first few recordings, but then I quickly realized the weaknesses of the chain. The C414 was from the 70s with a not the greatest condition capsule, and it was just starting to sound somewhat wonky from what I remember it being, and it was leading to dissatisfaction in what I was getting. The preamp wasn’t helping either, the API was great at giving my voice a bit of a push forward in calls and adding some nice hotness and grit without breaking up, it really wasn’t the right sound for recording a keyboard I feel, and it was emphasizing the weirdness of the decaying capsule. The Lyra was something I forced myself to scale down to when I had to sell almost all my audio gear (I am still incredibly depressed, it was my entire life), I made myself get something that was a compact combined high quality DAC and ADC and that lead to exploring a lot of interfaces only to find the Lyra was the only one that I felt offered equal DAC and ADC quality at a level that I felt was sufficient (even if it was many leagues below what I had), so that’s what I determined was the best option, and it wasn’t the bottleneck in this situation. I will admit, I kinda made a bit of a heat of passion choice when I decided on the next mic for the channel, a modded U87i base with a Thiersch reskinned capsule, higher gain FET mods with replaced caps and resisters/diodes, and an oversized Cinemag custom transformer, a fun project to try and get the sound of a good condition/sn vintage 87 without paying full good condition/sn vintage 87 prices. I do think it accomplished that goal, but truth be told while the 87 is easily one of the better options for recording mostly anything, it wasn’t the most true to life option I could’ve picked which was one of my goals when starting this channel. But it sounded good and performed well, but I just wanted to elevate things further. So I splurged a bit on the Josephson C716 so I never have to worry about it again, and that was the right choice to improve my recording quality, but the wrong choice in preserving my sanity lol. In the past I had a C700 and that was incredible and the C716 sounds like a cardioid successor, but my mind decided to ignore how needy that mic is compared to everything else, and I feel as though my recording quality has both improved and reduced now that there’s more problems because of that. But oh well, I will continue to improve things where I can and continue to take advantage of it. The M160 is there not because it’s good for this application, but because it’s an interesting comparison point to something more proper and served that role well, check my schizo rambling for more information on that and how it also exposed other weaknesses of my setup. The preamp decision was also equally important and somewhat of a heart over mind choice, but this one was a good option that I didn’t regret and works amazingly with what I have, a Crane Song Flamingo.1. I’ve never had a bad experience with their gear so it only makes sense, it fit the sonic role, and haven spoken with Dave Hill and really getting behind what he did, I felt compelled to have another CS product back in, loved his DAC/ADCs and compressors in the past, wanted to give one of his dedicated preamps a shot. They’re neutral, clean, organic, but have a bit of unique character to them that’s hard to put my finger on that just makes things sound more enjoyable without feeling like overt coloration. It’s very nice. For the preamp I was really debating between a Forssell, NPNG, Gordon, and Pueblo (which I will spare you the comparison between them), but because I wanted to put together a chain that I have more emotional connection to (because I don’t do anything professionally anymore), rather than just tools, I took this path instead, and I can’t really be happier how it came out. I am not happy how everything else besides the electronic equipment has become the bottleneck, but I am currently slowly crossing that bridge as we speak lol. Josephson C716 → Crane Song Flamingo.1 → Prism Lyra 2 / Neumann U87i (Thiersch Reskin) → Flamingo → Lyra / Nikon D5500 with Mogami Gold cabling, Altor HISO.
Now that that’s over…….. I’m sorry lol
When getting everything in and wiring it up, as a reminder, place things in a way that minimize any sort of interference, both electrical and acoustic. Don’t place your equipment in a place that’s more prone to RF or electrical noise (keep away from computers, monitors, other electronics), also maybe consider separating via circuit for electronics if applicable as mentioned in the power section. Also keep your gear out of the way from the recording space in general if possible, to avoid picking up any potential electrical/mechanical noise that might come from the equipment (if applicable) but also not introduce objects that might obscure or cause reflections in your recording space (though this is only really a concern with bigger equipment). Proper cable routing should be taken to minimize any sort of interference but also isolated from the desk if possible as vibrations can travel through cables into the mic, so be mindful of that. After this, it’s all down to setting up the actual recording environment, and understanding the technique to maximize your equipment, and then making sure to try and preserve as much quality when capturing and publishing your recording.
Recording Technique & Environment:
Getting a good recording requires a LOT of trial and error, and the technique to get there is so hyper specific to your space and individualistic to your preferences + expectations that the only way to know is to try and see what you get. There are established techniques and recommendations, what’s considered the “proper” way to go about it, but the best recording engineers I’ve personally ever met have flown entirely by feel, going in with an open mind and open ears to get the sound they want. Most people typically just think of recording technique as controlling the placement and positioning of a microphone, but in reality it’s sculpting your environment and understanding how to capture that environment with the equipment on hand. When you’re typing on and using a board, you’re hearing more than just the board, you’re indirectly hearing all the things that physically interact with the board where vibrations transfer, and you’re hearing the sound of the board and how the sound of it reflects and interacts with anything you have within the room either directly or indirectly, along with the room itself. You do not type in an anechoic chamber, the sound of a board in real use can vary significantly from setup to setup and room to room, especially in the lower level nuances. I think there’s 2 minds to approach this fact, either to try and isolate the board from the room through various means, or try and faithfully capture the sound of the board and the sound of the room as you would hear in person. Personally, I favor the latter over the former, as I feel as though it offers a more true to life sound than one would get attempting to capture only the board; while everyone’s space and setup is different and that makes direct comparisons from a recording in one inapplicable to the sound in another, I still feel it offers more valuable information if given proper context than the sound of a board with a room more isolated as that’s just not how we use keyboards so it sounds more unrealistic to me. It also makes it hard to judge how well you’re capturing a board if you have no reference to what it really should sound like, it’s easier in my mind to try and capture what you hear, rather than flying blind and getting a result that’s not aligned with irl experience (ends up being a more creative choice than anything as at that point you’re more likely to sculpt sound to preference). I think the most immediately important and easy to change thing to start with is microphone placement and positioning, so let’s start there
– Microphone Placement & Positioning:
Personally, I feel the most accurate sounding placement is close to where you actually sit and type is. I think this best represents the way that sound reflects and reverberates, how it’s shifted by the angle of the board and how sound interacts with the switches and caps and reflects around your desk, how it’s like to actually use a keyboard. Now, this doesn’t mean placing a mic in the exact place your head would be and calling it a day. Microphones do not capture sound like a human ear does (with a few exceptions, and yes I know what some might be thinking right now, binaural is cool but is not easy to get right), and most importantly they do not have a brain attached to them that filters out and emphasizes specific sounds as our ears do, you will need to do the work on your own to figure out a position that’s a good balance of “from the listener” yet minimizing and compensating for some of the characteristics of the microphone itself. You can’t expect that your first recording will sound just like it does to you in person, you will need to experiment until you get closer to that goal. That could be finding the right distance that makes sure the board is properly encompassed within the polar pattern and how the off axis coloration can shift the sound, or playing with proximity effect depending on microphone to use it to your advantage and record something that’s not overly limp or overly boomy, playing with positioning to deal with reflections around the keyboard and what should and shouldn’t be captured to get a realistic result, testing out the differences between angling in terms of on and off axis and where to actually center the capsule, and so on. A microphone generally will capture all sounds evenly, you have to find the sweet spot where the capture aligns with how you perceive it when typing, and really the only way to do that is to experiment and find out, there’s not much I can really write when it comes to finding that sweet spot. I could tell you that I’ve had good luck starting my placement by sitting at normal typing position, taking my microphone and placing it’s capsule level with my chin in front of my face, then angling the capsule down to center around the TY area of the keyboard and adjusting angle and height to taste, but that’s only with the Josephson, for the Neumann I found that having it a bit higher and more aggressively angled down yielded a more accurate result, it really will depend on the design of your microphone and your space for what is and isn’t acceptable.
Many choose to just record directly overhead as it’s the most simple and most immediately striking result, maybe a ft or just about above the board itself just directly on axis, and while this can effectively capture some of the sound of a keyboard, I’d argue it’s not a realistic/convincing capture. The first problem is that the way a microphone captures a source is mostly sound agnostic, as in it will largely capture loudest sounds the loudest even if that’s not what something actually sounds like, you’ll typically hear more direct case resonance and switch and cap interaction and with how the plate interacts with the board itself, but in a way that’s shifted from how the actual response sounds to our ears. You’ll hear way more low level interaction that you just wouldn’t when actually typing on a keyboard, and while the first assumption is that it’s a good thing and a more “accurate” recording, the difference in levels and sounds will change the way your brain will overall perceive the balance of sound of the board and lend itself to an unrealistic result in person. It’ll also bring proximity effect more into play which may unnaturally elevate low end and change characteristics with distortion. Secondly, you’re missing out on the natural decay and diffusion of sound this way, you’re not properly capturing desk interaction and how the sound is sculpted by the room/space/environment, and having tonal and timbral shift from real life experience, which is very important and can go a long ways into making a recording of a keyboard actually sound convincing. Finally, the way that sound interacts within a keyboard will lend itself to different sounds and results at different angles with how sound from switches bounces around under keycaps, how the sound of a plate vibrating and reverberating changes, how the case itself directs it’s energy and how that dissipates and so on. I think the easiest example to why I don’t feel this is accurate is a simple real world test: type on your keyboard as you would on your desk, listen to how it sounds. Next, pick up your keyboard and hold it as you would hold a cell phone taking a call (easier with a 60% lol), a few inches away from your ear, and press some keys with your other hand and hear how that sounds. Aside from hearing more lower level noise, you’ll notice the shifts in the other aforementioned aspects, and this is how your microphone will hear it too, except even more raw and unprocessed than our own hearing in person. This is mainly why I feel this isn’t really the best primary way to capture the true sound of a keyboard. For myself, I’ve started doing this in my own videos as a secondary, since it can be helpful to get a picture of the direct switch interaction and how the board is with less room involvement, but only as a secondary where I have my primary mic positioned in a more organic and natural way doing the heavy lifting, it’s just for fun and to maybe make comparing my recordings a bit more applicable to others. Direct overhead isn’t what I’d consider the absolute worst way to position a mic, and it can give some pleasing and occasionally useful results depending on your preferences and board, but I’d argue that it’s just not the way forward if your goal is trying to portray a realistic representation of a keyboard in use.
If your goal is realistic representation, why not stereo? In theory, it sounds like a great idea, we hear in stereo, why not capture in stereo as capturing space is infinitely better in stereo right? In reality, there’s a lot of complications to proper stereo capture at a desk that I feel it’s best to just stick with a really good mono source if you aren’t confident. To be honest to our ears at a desk, a keyboard is mostly a mono source, and there’s so many things in front of your desk that can cause issues with a stereo pair from comb filtering to early reflections to messy frequency and time domain, and just really finicky positioning, I’d rather go mono since that does the job well with less variables that can mar quality. It’s also a cost for quality thing as well, it practically doubles your cost in terms of both microphones (specifically a matched pair as well) and preamp/interface, so I’d rather put that money into a single better mic and preamp/interface instead. I also think that generally, personally, I find that stereo recordings also end up being a bit less intelligibly if that makes sense, it’s just harder to really wrap your head around what you’re hearing and the true tonality and sound of the board because of that (entirely depends on the recording though, this is harder with poor stereo recordings). But if you want to give it a shot, go ahead, just be prepared to put in a lot of effort to get usable results.
I would first avoid a cheap XY capture from something like a Zoom recorder, it isn’t what I’d consider accurate for this use case, it’s really not meant for this sort of application and is just too narrow and undefined with it’s stereo positioning, and just doesn’t sound realistic for this because of it’s capsules design. With a more proper XY setup using SDC you could make this better, although I still personally don’t think the sound of the results are convincing for something this close with most microphones in XY (you generally end up with a more vague center image and that just doesn’t sound right with a keyboard). An AB pair would be an imo better way to do it at a desk from a distance, because of that stronger and more detailed center, but also still requires careful configuration to avoid some of the downsides (it would need to be at the proper distance away, you might consider a Jecklin/Schneider disk to mic closer together, and you’d need to be very careful about phase issues). I am currently experimenting with this so I will be continuing to update it as I learn more, but I actually think the most flexible and ideal way to record stereo for a keyboard would be using Mid-Side technique, for a few reasons. First, you still get a clean accurate mono capture of the keyboard at your disposal which is what I’d prefer in a pinch anyways, so that’s always a backup and something you can play first (MS recording is also means that if someone listens to the stereo track in mono, they get that perfect mono image as well in some cases). Second, you don’t need a matched pair of SDC, you can use mixed LDC which are my preference of microphone choice, you just need a cardioid mic and a figure 8 mic that aren’t too far off from eachother (identical mics only matters for larger scale MS recording, though keeping them the same makes matching gain easier), and I prefer to use a brighter leaning figure 8 with lower off axis coloration to compensate for LDC high frequency loss and directionality at distance in a room and to give the stereo image more clarity, a more accurate balanced mic for your center. Note, while the options I mentioned mic wise will work if you get some of the multipattern versions, you will want to prioritize a smaller capsule, single sided especially, for a closer to perfect balance as most LDC with 2 sided capsules will be slightly imbalanced in their fig 8 pattern, so YMMV depending on what you choose. You can place it at your listening position and adjust to your room, one of the potential downsides of MS stereo having a 180deg null is actually a positive in this case since that’s where you’re going to sit and type! Positionally within space this is close to an individual point source from where a typist would be, and you can control the stereo width with blending to match what you feel is right. In post, you’ll either create a manual matrix or use a plugin like Voxengo MSED to decode your M/S recording to stereo and make those adjustments. Overall, this is the most flexible and potentially forgiving way to record a keyboard in stereo while still yielding accurate results to my ears, and gives you the ability to adjust things in post to how you think it sounds in person. Now it is more complex to setup in terms of creating a matrix or decoding your mics to stereo (this is not as plug and play as an XY pair), I think the upsides outweigh the increase in complexity, and I would argue this is also less of a pain than AB pairs in terms of setup/configuration and results (but I might argue that a near perfect AB setup would maybe give a higher fidelity result). Binaural is the one thing I just haven’t really tested for this application whatsoever, I think there’s high potential, but good true binaural is not cheap and easier said than done, so it will be something I might explore later, but not right now and requires some completely different equipment that I haven’t discussed here (the cheap heads largely suck and most binaural are more made for measurements rather than actual recording fidelity, you would get the most realistic results putting microphones in your own ears actually fit for recording like a DPA 4560 with some processing, but again beyond the scope here, I will stick with midside for the time being myself lol). My current setup is not ideal for stereo at the moment, I’ve done a few tests that aren’t perfect, but it’s promising enough that I think I will be using MS going forward if I can, I just need to optimize my setup and workflow around it. I know I said all this stuff about mono recording, but this is still in the spirit of mono recording, and I don’t suggest everyone go this route. Note that most of this article was mostly written around mono recording, similar aspects still apply but there is more that will need to be done that may not be mentioned in full until I go through this again.
Isolating the desk from your microphone (and mic cable!) is very important when getting a clean capture, aside from reducing just obvious rumble or handling noise, vibrations can also cause timbral shifts and undesirable resonances/fr nastiness along with increased overall noise floor for sensitive mics (sometimes more low end as well). The angle and orientation of the mic itself can sometimes also increase or decrease the amount that vibrations that it can pick up depending on how the internal capsule is mounted/isolated in the microphone itself (or if you have an older mic with some looser bits like a transformer in there). That being said, I think with some mics having a slight bit of coupling can offer a valuable perspective/insight into desk interaction, so it can be worthwhile for a secondary mic or one that might suffer from being a bit anemic as well. If you think about it, most of us are coupled to the desk already when typing, unless you fully always float your arms or have fully separate rests I guess, so it’s not really the worst thing in the world, but with the more performance focused recording chain you have, the more detrimental coupling will be to the sound due to their sensitivity and lower level information pickup. Even if your microphone is on a floor stand, you may find that some vibration can still transfer from your desk into the floor and reach your mic stand so be mindful of that as well (something like a untreated wood floor or bare concrete can be troublesome).
I feel there’s more to be said, but I’m at a loss of words, it’s so dependent on your specific environment, needs, and preferences that I don’t know what else to say, just spend a lot of time getting this right in your own setup and it will reward you, you are doing yourself and your recordings a disservice if you’re not putting in the time and energy to find as best placement as you can. It can and does easily make or break a recording regardless of the quality of the equipment you’re using. A fantastic mic can sound like ass if positioned poorly, and a crappy one can be surprisingly serviceable if done right, investing in your recording chain is worthwhile but only if you’re willing to put in the effort to actually take advantage of the equipment.
– The Desk and the Mat (WIP):
Obviously the type of desk and the deskmat can significantly change the sound and feel of a board, but the placement of the deskmat on the desk along with the placement of the board on the desk and/or deskmat will also impact final sound, anything placed on the desk, and the way the desk couples into the floor and anything else it touches will impact how the final result is. Aside from all of this shifting the actual IRL sound of the board itself, it also shifts how sound might react when recording, you might notice less reflections when using a deskmat vs not depending on positioning of mic, what you have on the desk itself and where it’s placed in the room will also impact the sound, but I’ll get into that more in the next section, along with the aforementioned coupling traits.
I need to personally explore some of this more, as this is an area that is sorely undertested for me as I am too lazy to get a bunch of different materials to use as a desk for proper testing on the most ideal one, I actually use a shitty IKEA desk that’s cheap and nothing fancy, it sure sounds like that too, it’s the biggest limitation in my setup and adds more coloration moreso than anything else right now. It just amplifies everything and leads to a very hollow and untamed sound that isn’t great. I’ve just neglected to change it because I’m extremely lazy and stubborn lol. I have noticed that boards tended to feel and sound best on a dense solid wood desk with non pingy/resonant legs that are decoupled into a more isolated surface from my more limited experience with different desks, it will absorb some vibration but not all. I would highly suggest avoiding desks with built in filing cabinets or drawers since they tend to rattle and vibrate more, as do desks with legs with extra metal blackplates for cable management or rigidity since those tend to increase resonance. My desk goes directly into acoustic membrane covered concrete which isolates from the floor more than normal and reduces vibrations from elsewhere in or outside from the space (from uncontrollable external vibration in a shared building to the movement of a chair or vibration from a PC from creeping in), and can also shift the sound of your boards as well. Side note, please don’t put your PC on the same desk you record on, more on that in next section though.
I use an Aiglatson Studio Synchronize jacquard deskmat and that’s a bit thicker and slightly stiffer than the soft/smooth material version I have, it isolates the board while still transferring a bit into the desk, bit of a middleground between direct on desk and completely isolated with something super thick and soft. It does allow the footers to sink a slight bit into the mat which does end up with some boards having more deskmat contact than just on footers but not as much as the softer mat (important in changing sound and feel). I do debate swapping out for the more isolating softer mat, but it would lean more into the sound of the deskmat rather than the desk, and is more hit or miss depending on the board, and I want to keep my setup consistent as constantly swapping deskmats would degrade that goal for comparing the sound of boards with each other. I’ve tried other deskmats which have yielded different results, I just think the thicker but not insanely thick medium stiffness jacquard mats are a nice right in the middle option to the sound. I would be curious about running my setup with a cork or other nonstandard mat material, but then it would just be less applicable to the average setup so I’d rather not.
– Room, Reflections, & Pain:
In any recording environment, the room and it’s characteristics are very important to determining the final sound of a recording. Recording a keyboard is not any different. Our ears are great at filtering out these reflections day to day, but with a microphone and how it captures sound, they can become very apparent and very detrimental. I had already went over previously why I think capturing the room itself was important to begin with so I will focus on just how to properly handle the room and environment. There’s a lot of aspects to address here, so I think I’ll break them down into things on desk, things in room, and the room itself to start off.
Ideally, there shouldn’t be anything on your desk, you would get the best capture of a keyboard on an empty, sturdy, desk. However, that’s unrealistic to how most people use a keyboard. Most have their desk with a few monitors on it, probably the computer itself, could have some office supplies or documents/books, maybe some trinkets or figurines, in this hobby perhaps your audio interface or other peripheral equipment, you might have something else bolted to connected to that. I think the biggest thing you can do is get your PC off the desk, and as far away from your recording space as possible. It’s a big (or small) box of noise in both acoustic and electrical, as well as potentially a large source of early reflections depending on how you position it on your desk. Personally I prefer getting some longer USB and display cables (within reason) and finding another small desk or stand to put the computer on as far away as you can, as well as configuring your computer fans to be on a more silent profile if possible. Second most impactful would be the monitors, but realistically there’s not too much you can do about that without impacting daily usability. A monitor is a big source of early reflections that can cause general grossness in your recordings, especially with multi monitor setups that create a pseudo dish around your recording space (it’s even worse with curved displays), if you can get them farther away from the keyboard with some gaps in-between multiple monitors, it will improve the sound of your setup. To a much lesser degree the stands of the monitors can make an impact as well, so getting them on desk arms could be helpful but beware that those very arms could be a source of vibration and noise when recording as well if not properly isolated. Next I would move any sort of resonant, or non solid objects off your desk that are laying around to avoid those creating noise when typing as well. Sometimes having heavy solid equipment on the desk can be useful to help dampen it in a sense, but it just needs to not create noise itself in the process. For myself, one flaw in my setup was the amount of monitors + the amount of things on my desk, I made sure they didn’t make noise with vibration (and some actually helped to dampen the desk), but they do create a decent amount of early reflections that cause more reflections to be recorded than heard in person. I have taken some steps to address that by further refining microphone technique, but also by simply removing reflection sources closest to the keyboard and replacing it with something to help absorb and disperse reflections instead. This wasn’t some fancy absorber or diffuser, it was just a plush dog that took up the space under my monitor lol, but it did improve the final result enough to justify keeping it there vs the rackmount compressor (not in use) that was there just for looks instead. I could still improve in working to reduce the impact of the monitors on the desk, but there’s only so much I’m willing to do right now without decently compromising usability. Your goal should be to largely reduce the amount of things that will cause early reflections, and/or create noise themselves, if you can work to remove those, you will end up with cleaner, more convincing sounding recordings.
For things in the room, if you can treat your room for acoustics more professionally, I’d highly suggest doing that. I’d first start with bass traps in corners and some diy acoustic panels (not crappy tiny foam squares) on the walls in important reflection point locations (things you will have to measure in your room). If you have hardwood or concrete floors I’d address that immediately with thick rugs and anything to help cover that surface. If you’ve got a couch or a bed or a bookshelf filled with books or a rack of clothes out in the open, that’s actually a plus and will help to absorb and/or diffuse sound throughout your room. If you can move the desk itself closer to the center of the room, you can also get a cleaner recording because of that. These are just basic considerations, this will not become a proper guide on how to treat your room, there are much better and more competent sources for that information, along with modifying the other aspect of the room for the most ideal acoustics, a lot of it is far too involved for most to really bother with when it comes to recording a keyboard, but there’s still plenty of small DIY things you can do that can make big impacts, just do your research there. My recordings are done in a larger open room in a basement, with carpeted concrete floors and a partially treated ceiling with some acoustic focused ceiling tiles, and a more rectangular room layout with walls mostly covered with treatment or objects, so I may have less immediate room reflections than a smaller bedroom or office in exchange for more potential background noise. It will all highly depend on the physical characteristics of your space for what works best for you, Google is your friend here. Unfortunately regardless of the work you put in though, sometimes you won’t be able to get a really nice recording space, there’s only so much that can be done without redoing from scratch, so experiment in different rooms/spaces if you can.
– Environmental Annoyances:
You can largely control what’s inside your room and in your immediate area, but there’s only so much you can do about the room itself and what’s outside of it without flat out moving to a different location, or rebuilding your room to be more acoustically isolated, which is largely unreasonable and unfeasible for most lol. Uncontrollable noises that might creep into your recordings such as weather, fauna, other people and machines, the natural background noises of your environment can be a pain to work around. Trying to find the right time to record, such as monitoring your area for the quietest times during the day/night and trying to record then can be a worthwhile endeavor. I found that for myself the later the better in lowering chances of annoyances (although then I run into the sound of insects during the summer). I try and avoid anything in the afternoon as that’s when you’re most likely to have traffic noise, noise from things expanding and later contracting from the heat of the sun, noise over your power if you’re nearby anything industrial, and so on. Something related to that which most also don’t consider would be the actual air conditions of the room you’re recording in, specifically temperature and humidity. Temperature changes the speed of sound, with hotter temps causing faster transmission vs cooler being slower, which does cause a real world difference in how things can sound. Humidity also changes speed of sound to a much lesser degree, but the bigger difference with humidity is that higher humidity increases higher frequency absorption. In real world impact, a humid space can cause a room to sound more dulled with a colder space sounding more lively, a warmer space offering faster less distorted time domain with a cooler one offering more noticeable reflections and echo. Different materials in a room will also change their absorption/diffusion characteristics depending as well. Ideally, a moderate temperature with a lower humidity is best. For me, living in a very hot and humid place at times, I do find this to impact the sound of the room which gets recorded alongside the board, but it also directly impacts the board itself, though not for acoustic reasons only. That’s only because lube on switches (if you use any) will change depending on temperature and humidity and generally cooler and dryer is more preferable for that from my experience, so getting a proper balance in temp and humidity to balance both the acoustic and lubed switches side of things is worth experimenting with lol.
While it impacts the physical way sound waves travel and interact, temperature and humidity also impacts the sound of some sensitive equipment as well, with some microphone capsules sounding different because of humidity, it can sometimes cause higher frequency loss with condenser capsules and higher noise floor, potentially even a short or crackling in extreme cases! I won’t get too deep into this, but the capacitance and stored charge within a capsule can vary depending on the temp and humidity impacting the way molecules release energy with the more humid environment leading to charge to escape easier than intended. It’s a small chance of actually causing a problem, and not a big concern in a regulated environment, but can be a problem for humid areas of the world. Your interface and specifically preamp also may have more ideal operating temperatures, a lot of high end equipment tends to need to warm up to sound it’s best (this is especially common with discrete class A designs which many quality mic preamps are), so letting them warm up for 30 minutes to an hour before recording is preferable, but impact may vary from design to design. I personally find that it takes my setup about an hour to fully warm up before it sounds its best. If I keep my environment very cool however, this time amount increases. But again all of this is sort of smaller potentially negligible depending on the conditions changes, but something to be mindful of in extremes or with specific environments and equipment.
You know what else is very sensitive audio equipment? Your ears, along with your sinuses. All of this also involves around how our ears themselves change how sound is perceived with these differences in temp/humidity/and pressure which is why sometimes your builds might sound different in one season vs another depending on climate and how that impacts you and your hearing, which is something you have to account for, so sometimes it can literally just be your hearing/perception more than anything else (and that can be more impactful than the above). When you’re planning to record and listen back to set things up, make sure to keep your sinuses and ears clean, not be under some sickness or wax buildup so you can get an accurate picture of what you’re doing. This really should be higher up in this article, but it just kinda fit well into this section and it is next to mic placement and desk information which are equally important, so it’ll stay here. After you’ve considered and experimented with (mostly) all that’s been discussed already, it’s finally time to actually talk about getting a good recording into your computer, and what to do with it from there.
Digital Capture, Editing, & Export:
Despite getting the proper equipment, configuration, and environment being the more resource and energy/time intensive, there’s still a lot that could be left on the table if the actual capture process and steps after aren’t taken with care. For myself, I found that being a bit more methodological throughout the process allowed me to make small tweaks at each step that added up to a noticeable improvement in the final result. There’s more that goes into getting good quality digital audio and processing than most realize from my experience.
– Initial PC Audio Optimization (semi-optional):
While a lot of modern computers and audio interfaces are largely plug and play in terms of getting sound in and out of the system, there’s still a lot to be done to actually optimize potential sound quality on both playback and recording within your computer. While just plugging in and calling it a day is good enough for most people, which is why I’m calling this step optional as it’s a lot of effort and potential hassle for a smaller improvement for most, it’s something that I’ve personally spent time doing and felt it made a meaningful contribution to a slight lift in my recording quality. You really better care about sound if you’re doing this, but considering you’ve made it this far, I think we might have similar goals :3c
If you want to do the bare minimum but ensure quality sound, make sure your proper interface drivers are installed (ensures proper exclusive mode support and allows you to tune buffer size and other metrics), make sure to set your interface sample rate at the highest rate you can in 24 bit space without experiencing buffer overruns/clicks pops (generally most actually high quality interfaces will sound best at their highest advertised rate, but play around and see what sounds better to you), make sure your digital volume is always 100% and not over or under for both input and output level (digital volume degrades sound quality), disable any sound enhancements/DSP on your system (obvious reasons), set your system to run on performance power mode (to reduce chance of xruns and have stable clocking), and keep running processes to a minimum when recording (same reason). Doing that will be good enough for most people.
I am too lazy to fully explain the why of everything for an optional section that most won’t touch, especially when it’s going to end up being more about Linux than Windows since that’s what I have more experience deeply optimizing for audio performance (it can and should be done on windows, but it’s a bit more surface level as there’s less you can really deeply tweak). It might be easiest to start with my own setup and how I built it with an audio focus.
Before US tariffs hit, I panic bought a new computer to build before things became overpriced or hard to find, I needed a computer upgrade anyways from a crappy Dell fleet laptop that I had been using for awhile as I don’t really do anything intensive on my computer anymore, and had dedicated hardware audio recorders and streamers for playback when I wanted a quality source. Now I wanted one box to do largely everything well, so I needed to be more meticulous when selecting parts for my PC build. My first priority was getting the best power supply I could with the lowest noise and cleanest power for lowest chance of noise issues in the system (went with a Titanium rated Super Flower who’s the OEM for the top tier EVGA PSU of the past). From there, I choose an 9800x3d CPU for a few reasons, it has pretty awesome single core performance for the money which is important for most audio tasks, second is that it’s the highest performing AMD chip that only has one CCD which the rest had 2, and that would be detrimental to overall lower level system latency for audio in comparison, and picked a large air tower cooler to reduce noise but also not have extra electrical noise from a pump in the system. I made sure to pick a motherboard with no RGB and no real gimmicks with quality VRM and high layer PCB for lower electrical noise, went with RAM that ran 6000mhz only dual channel 32gb dimms to have a good balance of capacity but also lower demand on the CPU and motherboard for less electrical noise, went with a PCI 4 NVME instead of 5 and only a single unit for lower noise as well. All parts that I felt were the least intrusive and lead to lower noise and lower latency than normal. I also picked a fairly quiet case with large slower moving fans to reduce any audible noise, although admittedly I have not tuned this yet and could make it more quiet during recording lol. Other physical modifications or things I did were to individually measure each USB port on each controller to determine the lowest electrical noise and then use that for my audio interface. I was also very careful with internal cable routing in keeping PSU cables away from everything else where possible.
From there, it was all tuning, let’s start with the BIOS tuning. I don’t remember the exact names for everything that I did, so this might be a bit vague lol, I will update this at a later date with more specific info. The first step I took was disabling anything in the system that I did not use, such as network devices like NICs, internal onboard audio, internal GPU, motherboard features like AI nonsense. From there, I started on tuning the CPU itself, forcing a steady manual underclock and undervolt, disabling any sort of power saving or sleep settings for cores, tuning cache and infinity fabric settings, then turning my attention to the RAM and manually tweaking cycles and voltage and timings, messing with the system bridge config, bumped PCI speed lower, disabled virtualization options and security mitigations, and whatever I could to get the most clock consistent and lean running system I could while still having a usable PC for some light gaming (that’s why hyperthreading/multithreading got to live lol)
Now onto the OS, I’ve been using MX Linux and some of the important changes and tweaks I’ve made there have involved modifying my kernel for realtime optimization like threaded irq and prempt rt along with increased freq for high resolution timers and tickless support, forced performance governor, specific audio user groups with full rtprio and unlimited memlock, manually configured process priories and core affinity to isolate audio to one side of the die vs the rest, messed with system swappiness and manual irq assignment, and so on lol. Also diving headfirst into very anal ALSA and Pipewire optimization to set the most optimal rates and configuration (clock, buffer, memlock, quantum, period, headroom, batch tweaks, resampling quality, etc), and force exclusive when needed. I also configured my system into a pseudo “Recording Mode” where specific processes are killed, devices are either handicapped or disabled, and only my recording software can run. There’s more I’m forgetting, but I do think that the small changes and tweaks add up for audible improvement in both playback and recording.
– Gain, Levels & Capture (in both analog & digital):
Gain staging is extremely important, trying to get the most optimal levels in analog and then only slightly boost in digital is what I feel results in the best sound. I think some people miss the real focus with this being overly obsessed about gain, but in the wrong way. While I could post my exact gain settings on my preamp, interface, and so on that’s only really going to be useful for my own setup, and not meaningful or insightful to anyone else, I also am not constantly changing the gain of my hardware either, only slightly boosting in post as I’ve already set optimal levels on my hardware. Aside from just being louder/quieter, something else to consider is that setting gain (for both mic and line gain) can impact the overall fidelity, for example your preamp may sound better at a specific mic gain range because of the circuit design itself (things like how hard you’re driving the stage, transformer saturation, so on), the potentiometers (specific sweet spots for best output for the circuit), and the preference in gain level for what you’re feeding (I notice that feeding my interface with lower gain or dropping the gain level on it will yield a thinner duller sound, going too hot will yield something a bit too forced and aggressive, with either option still avoiding any clear distortion or clipping). Trying to ideally match everything in theory might not yield the ideal results in practice, it just requires a lot of experimentation and trial and error, and you’re just going to have to play with your own setup to see what results in the most accurate sound to you.
I try to not boost volume/levels in post outside of a slight bump because it boosts the volume of everything, including background and quantization noise, so getting a hotter analog input signal into your interface can sometimes increase your overall levels without boosting noise floor as much so that’s more desirable to do (if you have good enough hardware that can handle it without it’s own noise), you get higher SNR that way and generally a better sounding signal, although you may encounter more coloration from aforementioned analog circuit quirks vs digital gain. I just worked to find a balance of getting a reasonably hot input without distortion into my converter and only boosting a bit in post which I think results in the best balance of volume and noise without hurting fidelity (digital gain also adds re-quantization artifacting; while it can be mitigated you still may need to dither again which isn’t great). Having better equipment helps here, if there’s more lower level detail to be had natively in the source, you can get more out of boosting digital levels with less issues too.
I would avoid using any sort of Pad or High Pass filter on your microphones, generally pad switches will reduce fidelity and will not be needed for this level of volume in recording (most often the capsule is still overloaded/distorting anyways if you have to use this due to SPL level), and while a bass rolloff switch might be helpful for recording a voice, there is enough frequency information in a keyboard within that realm that you don’t want to cut that out, even if your recording might seem a bit boomy. You can correct that boominess with better placement, but it’s also something that you need to figure out if it’s just a boomy board itself or a mic/recording issue before trying to get rid of it.
When you’re doing all of your test recordings or listening to live feed when setting up your microphones and seeing how your results sound, remember to give a quick check using multiple different pairs of headphones/iems/speakers if possible so you can get a more accurate averaged picture how what you’re doing truly sounds, you may find you get a more generally accurate sounding result over a wider range of playback that way, and it may be easier to catch errors on one transducer vs another, but I’d still prioritize/trust your best chain the most.
Ever since I started recording with video included, I needed to care about syncing audio. In comes the infamous mouse click “reference.” Some say a sound demo is worthless without, some say it’s useless to include instead, what’s the big deal? Personally, I’m sort of in the middle about it. I think it’s most useful as some sharp sounds to sync video to, but also a combined honesty test of how realistic the mouse click sounds within the recording setup and space, and maybe gain some shared reference off of that. Of course all mice sound different, but it’s still something someone could test at home if they feel the need to. I’ve seen some use other common “around the house” objects for this, that’s fine too, I just don’t feel like keeping a pen handy near my desk to click each recording, a mouse works fine and is a sharper click for sync anyways. You could make it be a source of level matching for both editing and also for the end listener (like “adjust your volume so the click and background noise sounds like x”) but I think the value from that is lost considering the lack of control of the rest of the listening environment, people will just adjust volume to whatever they like and their listening systems will have varying levels of quality at varying levels of volume, so I don’t really know if that’s worth bothering with lol. I just think it’s something to use as a reference to hear the tone and space of the environment and also as a bit of an honesty check. Everyone else does it so I figure I might as well too lol, although I will try and make sure to do it in a more consistent spot from now on lol. It’s a g502 for the 3 people who want an exact click reference at home I guess, not gonna overthink this.
A big thing as well is how people type during a test, I have seen some people barely touch their boards and that just feels so unrealistic and feels intentional to avoid hearing nastiness of a questionable build, or those who just absolutely hammer on a board for the same reason. Personally, I’d just type normally as you would with the keyboard if you can even if it might be a bit awkward with recording gear in front of you for the most realistic result. You can always switch to heavier and softer or faster and slower typing throughout your demo if you would like (I sort of do this myself, but largely just type as I normally do)
When recording, make sure to record at the native sample rate and bit depth of your interface, and use an audio method that’s either exclusive so your interface can only be accessed by the recording software, or as close to exclusive as possible by closing all other software that might have audio connections to the interface. Personally on linux I use Reaper to record the raw 2 channels, then export each channel as an individual 24bit/192khz (my native rate) mono lossless file (wav/flac) otherwise untouched, without closing the project. I then I move onto editing.
– Editing, Finalizing & Uploading:
If all you plan to do is hold that recording to listen to on your own setup and not much more, you can call it quits here. But most people are likely going to try and either share the recording in a more accessible lossy file, or they’re going to try and upload it to various platforms like YouTube or Soundcloud or whatever. There’s a decent amount of potential loss that can occur here if you don’t take care in editing losslessly and taking steps to try and minimize any losses that occur through the various platforms and formats.
One of the reasons why I suggested keeping your audio capture software open is because if you want to render for a specific platform, you’ll want to know what levels to meet to get optimal sound quality, and avoid as much degradation and penalties as possible to your recording. I use YouTube, and the levels you want to meet are loudness not exceeding -14 LUFS (which you can meter) to avoid getting hit with dynamic loudness control (also called stable volume), and -1db max peak in order to preserve quality and not get some parts squashed by a crappy lossly encode. Sometimes you will run into recordings where one specific moment of the recording is preventing you from boosting the volume without going over -1db, in cases where it’s only one significant event, I think it is allowable to carefully go in and reduce the volume of that section to a normal level (making sure to adjust only the specific acoustic event, this may result in light fidelity loss if done improperly), however if these spikes are a normal characteristic of the board you are recording, I feel it is more accurate to leave them if the board commonly exhibits sharp peaks/snaps from switch and cap interaction.
When doing any sort of editing, such as cutting or splicing audio clips together, or combining that lossless audio with video, make sure your choice of editor is capable of losslessly editing the files. I personally create audio cuts once I figure out times for the video and then export each file and assemble them in kdenlive to ensure no quality loss during editing, and be careful to not apply any sort of audio effects which would make them not lossless anymore. Really make sure your rendering settings for video are actually lossless, most often they aren’t, and you may need to make deeper tweaks to enforce this. The only edits I would personally make to preserve the highest amount of quality would be digital levels in the recording environment, and cuts to swap to different microphones or end a clip. I would not apply any other DSP or edits to the audio for both fidelity and accuracy reasons.
Post processing is a decisive thing in this hobby, and it’s more prominent than I think it should be to be honest. If you’re going to use it, at least disclose that. The main 3 things that a lot of people use to modify sound demos are compression, eq, and mic/placement. I already covered mic and somewhat placement previously (in the ways of overly colored or inaccurate microphones or recording techniques), but compression is really something that some people love to tack on, when it really should be used extremely sparingly, if at all. It might make your board sound more nuanced and dense as it pulls up more lower level information and smooths out the top, or however you’d like to spin it, but it’s not realistic, you’re artificially limiting dynamic range and forcefully bringing lower level sounds to the forefront by doing it, which just won’t map to hearing the board irl in use. It also can blunt the overall transients and actually reduce low level nuance if overdone or a poor fidelity compressor is used, and leave the board sounding homogeneous and dull. If it sounds good or not is subjective, but I wouldn’t personally consider that pleasing or accurate especially if it’s heavy handed or using a more character compressor instead of a cleaner one (compression itself will introduce artifacting and signal degradation, a character comp like a 1176 or la-2a will add more coloration just by existing, won’t get too deep into this though). The only compression I’m personally OK with is using one properly setup as a limiter in your hardware chain to skim a bit off the top when needed to stop clipping so you can turn your gain up a bit more and get a more usable signal. But only very carefully/sparingly, and you could harm fidelity a lot here if done incorrectly, a truly good limiter is hard to come by anyways. One might argue that using it with the right microphone can be carefully configured to sound closer to what you hear, fair enough, you just better really know what you’re doing and be honest about it lol. I feel the same way about EQ, sure you can bump the bass on your board or dip the treble or something and make it sound deeper or cleaner or just whatever how you want, but in most cases you’re better off not having it from a transparency/fidelity (due to the way it can degrade and distort audio) and an honesty perspective. If you really wanna justify using it to make it sound closer to ear, just be really careful. I would only consider these as a last resort, best not used, just leave it as is and at least get a more faithful in one aspect or another capture, but it’s all subjective. Oh and a nitpick, DO NOT USE A NOISE GATE OR NOISE REMOVAL/SUPPRESSION/CANCELLATION. You just ruin the fidelity of your signal by destroying low level info and the rest of it for that matter, sounds like ass and I’d rather just hear the noise instead of digital artificating, sthap. Hearing the room is a good thing anyways since it gives context. Fix your recording environment and equipment instead of doing this if you really want to improve, these are sloppy bandaid patches that will fall off sooner than you need them to.
For the final export to upload to YouTube, I spent a lot of time playing around with the formats, bit depth and sample rate, and aspects of the file, and found that rendering a video with at least 1080p resolution, with 1 channel (mono, not dual mono 2 channels render which many editors default to) lossless flac audio at 24/192 or 24/96 was most ideal in preserving quality after uploading. I thought it mattered in some cases even with keyboards, so that’s how I’m going forward lol. The theory is that youtube encodes your audio at a set ish bitrate in it’s lossy algorithms, by feeding a mono audio lossless source, this gives the best chance of the most amount of audio being preserved since it’s putting the whole bitrate to a single channel. If render in stereo but with identical left + right, now that bitrate is somewhat split between those 2 tracks even if they are identical, so there’s less theoretical quality loss with feeding youtube a mono signal compared to a duplicated stereo track because it’s a single channel at higher bitrate vs 2 channels (of identical audio) at lower bitrates. In practice, it’s extremely subtle but is present depending on the complexity of the recording, I don’t fully know what youtube is doing behind the scenes so can’t predict that it’s not messing other things up here, and it’s already at a high enough bitrate that it might not matter in most cases (a keyboard isnt a super musically complex signal so it might be more impactful on different things). But it’s still potential quality loss on the table, and this only is worthwhile because it’s a mono source already, so I might as well lol. I had also initially gone about manually dithering my audio down to lossless 16bit 48khz using my own methods because all dithering is not equal, and having control over your dither can tune it to sound best to your ears, along with that rate being what YouTube delivers to you, but after retesting that theory and not really hearing appreciable benefit (and sometimes worse quality results than expected), I think keeping the native rate with no dithering and letting the only processing happen through YouTube resulted in the highest fidelity sound through YouTube.
There’s only so good a lossy file can sound through whatever platform, lossless will always be better, especially to those with a high quality listening setup, but at the end of the day you have to accept that most people listening to your demo will not be listening on high quality setups, in a huge variety of environments, and I think there’s some worth in discussing how one might improve value in listening to demos.
Playback & Listening:
You can take all the steps you can in order to craft the best sound demo you can, but there’s still 1* final barrier to quality for both you and other listeners, actually listening back to it lol. It’s actually not just 1 barrier, there’s more than most realize, and they can become larger bottlenecks than expected if configured incorrectly.
The actual source, is it coming from YouTube on a PC? If so, you need to watch in the highest resolution video setting you can so you can get the best audio rates, and you also need to make sure any sort of stable volume is disabled if you even can with the actual video volume maxed, you also want to make sure you don’t have any settings/configuration on your browser that could either modify or reduce sound quality (uncommon but there are some). I can’t speak to the YouTube app since I don’t use that. If the source is a lossless file, then you want to use playback software that keeps the file bit perfect, and ideally exclusive mode, or some sort of higher than normal quality output like WASAPI on Windows or Pipewire/ALSA direct on Linux.
From there, you’ll want to make sure overall system volume is maxed so you control through hardware if using external audio equipment, along with having made some of the tweaks in the PC optimization section if possible in order to ensure there’s no DSP, bit depth and sample rate are set correctly, and so on. If you’re stuck with a crappy internal soundcard, that’s unfortunate, but you work with what you’ve got lol. If you’re stuck with bluetooth/wireless, that’s also unfortunate, make sure you’re using the best quality codec you can with system volume maxed and control volume through the headphone/iem itself if possible. Internal soundcards typically aren’t of sufficient quality for a high quality pair of headphones, inears, or line out to speakers, and you’ll find when pushed they can become harsh or brittle, so finding the right volume range where you’re able to offset the cost of digital volume to meet the best compromise. Bluetooth is worse being inherently compressed and lossy transmission, along with any sort of extra processing and DSP going on within the headset (and typically lackluster integrated dac/amp). So your best bet for quality is an external DAC and amp with a nice pair of headphones or speakers to offset a lot of the nasty. You’ll want to use multiple different transducers and equipment chains if possible to get a better picture of the sound and how your own equipment colors things if you really want to help minimize this (the higher quality the better). Not to shill, but since we’re on this site, if you’re interested in a shortlist of audio equipment I personally thought was worthwhile, you can check that out here, it’s a bit outdated now though. I should really not use a forum for personal projects like this lol.
There’s the whole discussion about ears and expectations as well, everyone’s are different, and there’s only so much you can do to try and offset the way that shifts sound. Just something you have to understand and live with for the most part. That being said, hearing is innate, but listening is a skill you develop with time and experience, and the better listener you are, the more value you can extract from some demos if quality allows.
Despite all of this, what I would say the absolute most important aspect understanding a sound demo is getting context to go with it, which is why it’s so important to provide that when uploading a demo, you can allow the listener to listen to your demo with more information than they would have otherwise, that allows them to listen to it from new perspectives under new expectations and understanding, which just results in a more worthwhile experience for both the poster and the listener imo. You do have to consider that because you have no clue what the board actually sounds like in most cases, there’s only so much value you can take away from a recording. It will never be as useful as real life, should not and can not be used to replace real world listening experience. Don’t get too wrapped up in this (ironic coming from me), and learn to accept that most all recordings (mine included) are not perfect and will never be perfect, so use your best judgement in what to actually trust about them when you encounter one.
Random Extra Thoughts:
This will fill up with things later as I think of them, but for now, I mean I have thoughts about this hobby overall, but I will keep it related to recording.
I think I already mentioned previously about how I don’t like the current state of recording boards, it has gotten better than the early years where people just whipped out their webcam and let it go, and obviously those who still do that at least now benefit from better webcam mics out of the box lol, but it still seems so neglected considering how passionate people are about boards and sharing them sometimes. I am just surprised to see some really large channels or prominent people in the hobby who already record boards often make such questionable decisions in my eyes lol. I understand that not everyone really cares about this as much as I do, but I think people should care a bit more in general lol. It almost feels like some have just given up without any real care toward what they do, others somewhat misguided based on trying to replicate already misguided demos, it ends up being a negative feedback loop where people try to make their demos sound like the flawed demos that already exist, and you just end up with more flawed demos lol. This also makes it hard to get actual good commentary on what makes a good sound demo since people are so pulled to what’s already well established, and sometimes actually don’t like the results of a more accurate sounding demo because it’s too different from their own expectations. I’m not sure, I just think there’s not enough reliable info out there, so that’s partially one of the thoughts I had to justify writing this in my head.
I am sure there are groups of people who would take significant issue with what I’ve said and what I suggest specifically relating to almost anything in here lol, I would assume largely those who believe that only measurements lead to true accuracy and fidelity which I don’t personally believe whatsoever after my journey through audio, but to each their own, this is only my experience and my suggestions based off my experience, and it’s easier to ignore those who don’t have real hands on experience with things telling me I’m doing things entirely the wrong way lol. It is what it is, I just go based on what provides me the best experience to my own ears and my own methods, through both casual and actual proper testing methodology. As I’ve already said before, you can go about this in an entirely different way from what I’ve said and still end up with a worthwhile result, this is only what I feel works best for me, and what works best for me may not work best for you.
Final Thoughts:
I really appreciate you reading this far, thank you lol, means a lot. I don’t really know what my true goal was with this was, it just became a big brain dump from having an imaginary conversation with myself and reflecting on what I’ve done so far, very normal and healthy behavior. I care too much about what I do sometimes, I should put this energy toward figuring out how to properly lube topre better instead of writing a giant wall of text that isn’t going to be of any use to 99.5% of people lol. I do hope that there was at least something in here that was either helpful to improving your own craft or at least somewhat interesting so it wasn’t a complete waste of your time lol. This will end up a living document for as long as I still record boards, as I end up constantly changing and refining what I do which will lead to new improvements and reflections on what I did before, I’ll try to keep this up to date on that and note down what’s changed. If you have any comments or suggestions or info that would be good to know about improving what I do, or just want to talk about stuff, just let me know, I want to explore all avenues I can in order to improve what I do, and I’m surely missing more than I think I am with stuff, there’s always more to learn lol. I won’t take more of your time, thank you, see ya o/
Postramble:
I ended up writing this in the span of a week, so I apologize for the poor formatting and potential errors, I am not one who excels at organization and making comprehensible writing lol. It’s hard to try and condense all of your thoughts and experiences, and then try to translate it to paper lol. I ended up writing this almost as if it was giving more advice than just random thoughts which I apologize for, it was more written for myself in a sense, if I was the reader specifically more than anything else. I will likely end up reformatting things to be more general lol.
If you want to see the original precursor schizo rambling comment threads, I’ve linked the first one here, and the second one here. This largely ends up covering most of what I’ve said in those (with more context in more detail, in some respects), but those are more personal to my setup and my reasoning behind my own setup than this article ended up being (more generalist and wider scope) so maybe there’s still something worth looking in there if you’re bored or want even more content of someone talking to himself. If you didn’t come here from my channel, might be worth giving a listen for some of the audible context behind what I write lol.
Mostly typed on a Pro1 or 86u, with various boards in between
Had the idea and rambling for a year or two, I don’t remember most of the timeline as it’s scattered throughout txt files, phone notes, youtube comments, and discussions with people, but better late than never lol
9/30/25 – RC1 – First soft release.
10/5/25 – RC2 – Had a radical shift in opinion with stereo recording, added more elaboration and considering shifting setup to suit Mid-Side more, also realized more limitations.
10/15/25 – RC3 – Cleaned up things a bit more, a bit more hesitation on what I’m doing and if it’s truly the way that I should be as I become somewhat dissatisfied with the quality of my recordings, but I think that’s largely environment and placement more than anything else, although I should experiment with a more proper AB pair and get a better Sides mic to give that a fair test again.
Not a reviewer and not writing reviews, just opinions and experiences of someone in too deep for fun. Prefer to keep things concise (sometimes), casual, and hopefully enjoyable, not much more to say here lol, more active on the forum
Add me on XMPP at m0nm0n@hookipa.net