Gain on an AI...how much is enough

[jimmys69]

I actually contemplated this situation myself and wondered what it was all about. Luckily I never owned a low output mic like the SM7b before I had preamps worthy to use it.

So there is obviously an issue with low gain mics and interfaces with low gain to run them without the noise floor and distortion being an issue.

I am learning what works here for others as well.

 

[joey2000]

I'd bet a preamp with 50db gain and no audible hiss when maxed out would be just fine for an sm7b but that's the gamble.

And I can't likely afford an Sm7 so that's a problem.

The vast majority of us aren't going to know until we plug it in.

So's that lol

Yeah, fair points though, it's not just how much gain but how "quiet" the preamps are. Which as you say you don't really know until you're there. And how good this or that AI is in that way is all opinions and they're like.......you know.

 

[garww]

Well, you can go out of your way to research the preamps I mentioned. A member here has a Rane , ms1b???.Low Dollars invested in it. The software noise cleaners do get a lot of use, though

 

[ecc83]

I don't have an SM7b (might see if I can hire one?) but I do have a very wimpy Reslo RB ribbon.

The mic has to be 50 years old at least so the magnet has weakened a bit and it is a 30 Ohm mic. It is boosted a bit by a £40 Sowter 1:4 transformer but that does not fix the main problem, hum. The RBs were not wired to buck it and therefore, as you can see from the RTA plot, lot of 50Hz and above.

The gain on the left mic input on the KA6 is at max and the right at min. You can see I think that the flat out noise is only about 6dB higher than the minimum pre amp noise? If it were not for the hum and its harmonics I could crank the level in Sam 8 even more. As it is, speaking at about 2", the S/N is borderline acceptable peaking to -20dBFS.

The 7b has really excellent hum bucking ability I understand?

Dave.

 

[ecc83]

Repeated the exercise with a 57. Some 50&100Hz but generally a lot cleaner and the mic is a wee bit more sensitive.

Hmm? Don't know where that hump comes from at around 84Hz? Be interested to see plots of other 57s?
Dave.

 

[joey2000]

If there's audible hiss it's not going to help recording at lower gain.
You still need to make the signal louder, relative to the hiss.

I get what you're saying (not just gain but "clean" preamps), but depending on specifics, it might, mightn't it? There might be a break point and below which you're good, cross that and you can hear noise you don't want to hear. In fact that's the gray area I'm concerned (paranoid?) about.

 

[Steenamaroo]

I don't think so, no.
Maybe on of the more tech guys can correct me but if hiss exists before the preampfification it's going to be amplified by the same amount as the voice.
In that circumstance, it won't matter where your gain is.

 

[garww]

The passive guitar pickup-jack and the active amp are full on 'til you attenuate.

So, if the music is full on, no one will hear circuit noise : )

Besides, without noise, one would have to go record a distant jungle waterfall to satisfy the ears aversion to silent space

 

[ecc83]

I don't think so, no.
Maybe on of the more tech guys can correct me but if hiss exists before the preampfification it's going to be amplified by the same amount as the voice.
In that circumstance, it won't matter where your gain is.

Not that I claim any great "tech" but I have lived through and had a lifelong working interest in the steady improvement in dynamic range.

Noise, the whole 'thing' is a movable feast, how annoying it is depends on many factors. The best vinyl could manage on replay was about a 70dB Dynamic Range but headroom was very tight. Tape, without NR was a bit worse, 65dB but much depended on how hard you wanted to hit it!

In a very quiet, domestic listening room vinyl (from tape of course) was only JUST satisfactory for modest DRs such as small classical forces. Even the grand piano reproducing the heavy Chopin and Beethoven scores was beyond the black disc and Engineers (proper ones mind!) had to ride the faders with great skill to get it all on! Most people in recording studios back in the disc heyday could read scores very well indeed.

Today the tables are turned. No longer is the 'storage system' the weak link. For $150 or so, ANYONE can have an AI with a -100dB noise floor and in fact for the vast majority of people it is their room and beyond that limits their noise floor.

The 'noisy guitar amp' was mentioned? Good example of how noise bothers/don't bother us! Down the Mucky Duck we bash out 100dB+ songs and care little for the hiss, hums and buzzes. Put the same amp in a bedroom and FFS! How can I make a decent recording with this pile of ***T!

Dave. (and I am very pleased to have had even the tiniest of parts in the very successful development of what is probably THE quietest 5watt guitar amp on the planet!)

 

[garww]

---------- Update ----------

As things change, my my, how they stay the same.

Here's a 20-year old PDF on Vinyl Dynamic Range;
https://www.analogplanet.com/images/vinyl-dynamic-range.pdf

Of course, many people bought the vinyl TELARC 1812, and then, the CD when they could afford a CD player.

People still buy the Analog noise, as with this analog drum machine;

 

[joey2000]

I don't think so, no.
Maybe on of the more tech guys can correct me but if hiss exists before the preampfification it's going to be amplified by the same amount as the voice.
In that circumstance, it won't matter where your gain is.

Yeah, but in at least some (and probably most) cases, what you're recording will be considerably louder than the hiss and so hiss at lower levels might not matter. I think. Course then maybe it wouldn't matter anyway.

 

[Steenamaroo]

You would think so, but then that brings us back to the guys who turn up here with 58s and cheap interfaces wondering why they can hear hiss.
Those threads are almost always V/O or podcast guys who have periods of silence where it's obvious, for what it's worth.

 

[joey2000]

In my case it's possible silent/quieter moments in songs......and the 58 isn't exactly a "cheap" mic. Inexpensive, but near as I can tell quality enough that it should not in itself present an issue.

 

[Steenamaroo]

Fair enough. I don't meant to make out that it's this big massive deal that will ruin everything you record.
If the question is can 5db make the difference the answer is, in general, the higher the gain the higher the quality and the lower the noise floor. I think*

 

[joey2000]

Hence my paranoia about AIs with a little less - sigh

 

[Steenamaroo]

It's worth remembering that it accumulates, too.
If you have preamps with hiss that you can just about notice, you'll notice it for sure when there are 10 tracks.

 

[joey2000]

?? The more tracks i have, the more things I have (more instruments etc) which should cover up any hiss, I'd think, even if there is more in the background from a tech perspective.

 

[Steenamaroo]

If your tracks permanently have loud audio on them, sure.
If, like in the real world, you have gaps, pauses, breaks, periods of silence, fades, etc, accumulative hiss is a problem.

It just means you're forced to a: put up with it, b: noise gate everything, or c: manually edit out any problem areas.

If you're thinking, psssh, that'll never happen, go ahead and buy a noisy preamp.

 

[garww]

And, there again, the gain spec doesn't mean much by itself. If you want something good, plan on paying for it. With the baby Golden age at $250, you are overpaying $100

 

[jimmys69]

And, there again, the gain spec doesn't mean much by itself. If you want something good, plan on paying for it. With the baby Golden age at $250, you are overpaying $100

Wait, what did that mean? Did I miss something? Was the 'Golden Age' preamp mentioned in this thread? Why was it over priced?

Sorry if it was...

But we still now have the discussion of running a preamp line level outs in to an AI with preamps that do not have line level inputs. Unless they have a line input, then you are dealing with a second stage of gain on the way in.

So my question to others is why that is an issue. I know that I do not have any issue with my gear, but I could see that possibly being a bad idea with a low end interface with only mic inputs.

I really am questioning here for others as I do not have the issue with my current setup. Actually didn't with my previous interface (TASCAM US1800) as it had separate line inputs. I now have UR824's that do not have actual 'line inputs'. But they do have a -26Db pad that seems to work fine with external preamps. Plus the interface preamps have lots of clean gain to begin with...


This is a discussion that has happened over and over. But it seems it is time for a new one. Products change, opinions change, members change...

Next!