Enjoy the benefits of a premium membership!
Upgrade Account
-
TalkBass has been independent since 1998. Add your voice.
Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
Join free Log in
Want zero display ads or expanded classifieds tools? Compare plans.
Wanting a Sterile Tone
- Thread starter epimetrius
- Start date
Basically get a DI (an active one, probably), plug into it before the amp, send its parallel out to the amp's input, then take both signals (the one from the amp's DI/mic and the one from the DI before the amp). Even a Sansamp (turned off, it should be a regular no-frills DI) will do the trick. I don't even think you need a top quality DI, both because you don't need a very nice tone but just a clean, uncoloured one (easy to do with, like, a dozen parts and good component choice). Transformers are revered for their tone qualities (very subtle distortion) and because they used to be much better at balancing a signal, but these days it's really easy to make a well balanced DI, there are lots of very good opamps that do the trick.
Many options. As most have stated , a DI that feeds the amp and also goes direct in to the interface should provide you everything you need (two separate bass tracks). A benefit of a direct signal recording is that you can always re-amp it later if you’re completely insane/fussy/particular/obsessive (like many players).
Insert any guitar practice amp here, preferably open back. I realize you alteady have something you're working with, but if that doesn't get it for you then like any of these might:
Guitar Combo Amps | Sweetwater
What @mcnach stated was correct. The preamps in the Scarlett series are remarkably flat given their price point. This isn’t speculative as it has been demonstrated repeatedly on various recording forums when people complain about the Scarlett preamps. The problem, for many folks, is the relatively low gain on tap for the preamps. If a full range signal is the goal the Scarlett preamps are about as good as it gets in the consumer-tier for flat response.
It's really not that hard to make a flat preamp. Once you design the frontend (to get a balanced signal and make the most of it), all you really need is an opamp afterwards. Most opamps are as close to ruler flat as it matters in the audio range. I have never found a real bottleneck in my Scarlett audio interface¹, and besides, the gain pot is there to be used fully, if required. Don't be scared about cranking it, technically it actually sounds best if you don't attenuate anything.
¹: well, of course it's probably upscaling the voltage because USB doesn't provide that much voltage to begin with, and that means that it probably can't compete with something that has much higher voltage available to begin with, but I've never bumped into any real limitations, such as low headroom or noise or whatever.
I don’t believe the Scarlett series has any function capable of attenuation aside from an input pad which may or may not result in better sound once the signal hits the AD stage.
Cranking the preamps is fine but comes with two caveats:
- Cranking a Scarlett preamp, or any preamp for that matter, can get into the self-noise territory of its performance. At 24-bits once you’re downstream of the AD stage you’re theoretically 144db above the noise floor in the box, so I would rather go in a little low and make up the gain needed to get to a working signal level digitally where it’s totally free if you’re getting into preamp self-noise conditions. Even on older models operating at 16-bits you’re still theoretically 96db above the noise floor.
- You’re not clipping the AD stage on the way in as a result of cranking the preamp.
What I'm getting from this thread is a lot of disagreement on the definition of "sterile". I this context I take it to mean "lacking artifacts or any overemphasis on specific frequency ranges". A lot of people will use "sterile" as a somewhat negative adjective, disliking a given bass or pre because it doesn't reproduce the sound they themselves find desirable in a bass.
That's what I got from it as well. My experience with Rumbles is exactly that; warm, round, and colored, and that's not something I want (at least, not baked in quite so much).
+1
You can only add gain digitally to what has been captured in the first place. I don't know exactly how many true bits is the AD, I'm guessing between 20 and 21. Then add analog self noise. Self noise is going to be there whatever you set the gain pot at (actually, lower at full gain tehnically). The vast majority of designs is fixed gain with an attenuator (with the attenuator being the gain pot). Of course, when I say you should crank the gain pot, always within normal levels, if you're clipping the AD there's no point in doing so. And in fact you should operate the preamp so that it's VU level is -18dBFS (which means, usually, peaks at -12 to -10dBFS, but no real problem going over that). The preamp may show more distortion than other more expensive preamps with peaks getting closer to 0dBFS (in essence, lower headroom) but that's very, very, very rarely going to be a problem with bass anyway.
Indeed, and being so high above the digital noise floor it's very easy to take a signal that only registered in the sub -20dbfs range and get it as high as you want without ever getting anywhere near the noise floor. This will almost always be less noisy than cranking a consumer-tier preamp to get every drop of gain possible in the analog domain.
Modern AD stages are so alarmingly good that you might be surprised, but admittedly I've never see a test to confirm. Regardless, at 6db per bit it's a complete non-issue.
Yep, there's no getting around the inherent self-noise of an analog system. Where I disagree with you is the notion of "lower noise at full gain." Consumer-tier preamps are notorious for generating excessive and disproportionate levels of self-noise when at full gain which, for the purposes of this conversation, has no real world definition besides "cranking the knob."
I'd go so far as to say that it's a complete non-issue in any design for any sound source for a preamp design that doesn't contain transformers in the audio path that can saturate. Most of these designs are entirely linear from top to bottom save for the most minimal quantity of measurable distortion. Paying less than $100 per channel for Focusrite's consumer-tier versus >$500 per channel for a comparable non-transformer-equipped linear design professional-tier product like a Grace is measurably a "game of inches" as they say.
But what about "soundstage"? You gotta pay big bucks to get that!
EDIT: Only half-kidding, actually. I work on high-end gear, and lesser gear as well. Sometimes it honestly does not matter much (in the right hands). But yes, there is a difference....
Last edited:
Don’t forget about linear preamps supposedly lacking vibe, mojo, and any other adjectives constantly used to denigrate designs that are not at least loosely based on a 1073 or a 312But what about "soundstage"? You gotta pay big bucks to get that!
This is actually a misconception. The best performance, SNR-wise and THD-wise, is or should be at 0VU. Audio gear, I'm going to assume the designers know what they're doing and they basically design the circuit as to give the best SNR and THD figures at 0VU, which works out at -18dBFS for a sine wave. Any less, you get more noise relative to the signal. Any more, you get more distortion.
It's more that, without going through tape and lots of analog electronics, and having basically at least 120dB digital resolution from most audio gear (not counting jitter and other things that may or may not be a problem), you don't have to painstakingly ensure you have decent levels coming into the interface. But preamps do still have a SNR figure (which is still almost always miles better than what you had with tape), and the best performance is had at 0VU.
The real problem is not cranking the knob, it's the preamp's self noise. A decent (simplified) design involves a front end (or a transformer), maybe a buffer (it depends on whether it's a discrete or an opamp design), then a first gain stage that brings the level up to a certain point, then a "gain pot", then finally another gain stage which actually brings the signal up to the desired level (+4dBu, or -10dBm for consumer electronics, at least once upon a time), once you have set the gain pot where it should be. The problem is that those first stages before the last one have their own noise, and if that noise is high to begin with, and the signal you're working with is very low in level, then you have to crank the gain pot in order to get it to the desired level but you're also bringing along the preamp's own noise together with it. But whether the preamp's self noise is very low (and it's quite low on the Scarlett AFAIR) or very high, if your signal is low in level, you're always gonna get a better, more workable signal if you crank the gain pot than if you add gain digitally afterwards.
Of course, up to 0dBFS I'll expect the Scarlett's preamp + ADC to show both good SNR and THD figures. But still, why is the recommendation to keep peaks at -10dB (ish) and general levels (textbook definition: levels over a 300ms integration time) at -18dBFS (or even -20dBFS, as some argue)? Because that's where everything works best.
On the other hand, expensive preamps either have stellar headroom and noise figures, so that it doesn't matter much how you set it and you'll have professional-grade performance at whatever level, high or low (before digital clipping), or they have very good overdrive behaviour so that they actually sound nice when you crank them. The Neve preamps are renowned, for instance, for being really nice when cranked (and maybe also pushing the output transformer and attenuating the signal afterwards). The fact that they have good input transformers also ensure that they are virtually immune to grounding issues while still having great low-frequency extension with little (and gradual-onset) distortion. Now (well, for at least a decade) they also make expensive converters with good "soft clipping" behaviour so that before clipping digitally you also get some nice saturation (misnomer) going on; it used to be a craze in the mastering industry a few years back. (To me, clipping the converter is madness. A good converter should be as transparent as can be. There are many saner ways to get colour).
You know, I had those GAP Pre73 pres that are basically a clone of a 1073 (with not as nice transformers, but still quite decent ones). I sold them after a while because I realised that, apart from the fact that I'm not sure I really like the Neve sound that much (even after using the real thing; I find them a bit "dusty" and woolly, especially when pushed a little, as if they lack sparkle up top; I like them on tom toms, and not much else), I find that I much prefer to record with "linear" preamps and get my character either ITB or before the preamp. The former is non-destructive and you also get a much wider choice (just record the damn thing well, and then choose whatever flavour works best, and fine-tune it, and maybe change your mind halfway through the mix - even if this is BAAAAAAD), the latter, for electric instruments for example, is much better because you will generally get to use those sounds in a band setting and you can really get a feel for it.Don’t forget about linear preamps supposedly lacking vibe, mojo, and any other adjectives constantly used to denigrate designs that are not at least loosely based on a 1073 or a 312
And anyway, the preamp's contribution to the sound is so minor compared to every other factor, that I'm not sure it's worth bothering to spend an arm and a leg on them. For the price of a fancy preamp you can get, like, eight channels of a very good preamp such as RME's stuff, or not bother at all and just get a Behringer ADAwhateveritscalled, if you have your own project studio and don't need the reliability and support of a big serious brand. The only piece of gear I'd spend money on is an Avalon pre for DI bass. Really, that just sounds lovely.
Last edited:
It's a tested conclusion, not a misconception.
Now here's an actual misconception. There's no real correlation between voltage in the analog domain and decibels from full scale in the digital domain. There are reference points accepted as common such as -12, -14 and -18, but they're not actually a true correlation since they can all be an option depending on the use case. You can calibrate your analog gear to whatever digital reference you want and it's entirely preferential to the end user and the particular gear in question.
Agreed. Way too much is made of "getting as much signal before clipping as possible," especially when totally linear circuits are being described and digital gain is free.
Of course they have SNR, and measurably in many, if not most consumer-tier products such as the Scarletts being referenced here the noise is appreciably higher relative to the signal when operating at or near maximum available gain.
I totally disagree and there are many measured results showing the opposite when it comes to cranking a consumer-tier preamp. Setting it within the idealized range of the preamp's gain structure where SNR is at its best will be good. Cranking it creates a disproportionate increase in noise relative to the signal compared to operating at more nominal levels.
This is my point. Don't run a consumer-tier preamp up to the end of its gain structure and you'll have demonstrably better SNR. I can't stress this enough: I'm not speculating. The Scarlett series in particular has been shown to have measurably worse SNR at the top of its available gain structure, and thus backing off the gain a little and compensating in the box where gain is free will likely yield a cleaner result.
The problem with that approach is you don't generate a reason to buy another piece of gear. C'mon, this is Talkbass, we don't come up with logical, fiscally conservative approaches EVER!!!
You don't need a DI. You interface already has the correct input impedance for your bass.
