How much of a difference will upgrading an audio interface make?

I find there is a bit of muddled thinking here IMHO? We are surely comparing converter quality not the ancillary parts such as pre amps.
Early AIs did indeed have poor pres, low in gain and high in noise. They are much better these days but, since most AIs only have a single gain control for the pre amp they are limited in the range between high gain and high headroom. 60dB is about the practical maximum and even that leads to the next problem... 'Gain Bunching.

This is the effect you find where the gain suddenly jumps up at the last few dgrs of the pot's rotation. Thus, gain pots are not linear in angular rotation and gain. Putting a knob 'half way' does not give you 'half gain'. Better pre amps use more expensive pots with special laws but the ultimate is digital gain control (see RME)

Nobody outside NASA is "saturating" a 57/58! The onset of distortion for the 58 is around 160dB SPL at 1kHz and the output voltage would be +12dBu or 3.2V rms, well above the headroom of most AI mic inputs.

And finally, to be REALLY pedantic, nothing is ever 'saturated'. The term means a device is turned hard on and ceases all linear function. For a transformer (or mic coil) this means its impedance has dropped to its DC resistance, for a transistor it is turned hard on as a switch and probably has a resistance of a few tens of Ohms. Valves are never saturated by design. It buggers them up.

"Saturation" is one of those terms that has entered the audio lexicon but what people really mean is progressive, harmonic distortion.

Dave (I'll get me coat!)
 
I got sidetracked talking about mics and saturation. I do realise the discussion was mainly on the AD/DA side. There is an improvement in sound quality from when I upgraded from the Line6 UX1 to the apogee duet, but that is not saying much. The Apogee was at the time a pretty high quality 2 channel interface, it cost a fair bit also. But it was worth it. I do think there is merit in talking about te pre-amps to an extent, afterall. a lot of us are mixing our own recordings.

Those are interesting numbers about the 57 and 58. I am not technically minded, you are probably correct but whatever the mic is doing to transients and the character it adds is not the same as I'd get from a (clean) mic. I didn't want to get too into it but I'm not running my mic clean into my I/O, I'll get a lot of the saturation from my external pre, and DI box, so it's not entirely that important if the mic itself is not saturating by itself. It's handy to know that whaling some guitar chords loud into the mic itself won't help me get anymore saturation from the mic itself though!

You are spot on about the pre-amps pot though. I do remember when I would get near the max then the gain would jump up significantly (along with the noise) in hindsight I think I would have got much better results by dialing back the pre-amp gain and recording a very low signal, then gaining up inside the DAW. I think I would have got better signal/noise ratio that way.... maybe

Cheers for the info, I am definitely glad you posted those specs.

Edit: Is there any chance of saturating a ribbon mic before you stuff it up? From what I understand ribbons saturate before dynamics, and dynamics before condensers?
 
Last edited:
This sounds a lot like when guitar players talk about how they can "dig in" to their pickups and add crunch. Yet if I plug that same guitar into a Fender Twin set on 1 or 2, there's no crunch in there... it's pure and clean, no matter how hard I hit it. The crunch comes when you slam the preamp or output so hard that they overdrive and distort. That's why so many have used a Fender Champ for recording crunchy guitars, like Joe Walsh did on Funk 49 back in the day. You can run the thing wide open, so it just distorts like crazy.

Jam, unless you have a really quiet room (and I mean REALLY QUIET), you probably aren't going to hear much difference between gain from the preamp, and gain from the DAW unless it's a really CRAPPY preamp. I would bet that if you are like most of us at home, the ambient noise level is probably about 50dB below your loudest vocals. Shouting is really about 90dB or so. Set up a mic and set the level so that your loudest vocal hits 0dB, and then look at the bottom for the dancing noise level. What level is that? Guaranteed that it's not mostly from the preamp or mic.

What you should find with the 57 vs a condenser is that the transient response is much slower. There's simply more mass to a dynamic mic diaphragm than a condenser's. That means it takes longer for things to move. Compare a 12 or 15" woofer to an electrostatic or planer speaker. Big difference in transient and frequency response. That's the reason that big woofers aren't full range. There's too much mass to react to small fast variations in the signal.

Ribbon mics can have excursions that are simply too large to avoid damaging the foil ribbon. This is especially true of plosives and if you put in front of the port of a kick drum. Lots or air movement, which damages the ribbon. A passive AEA R44 ribbon can handle 165dB before you get to 1% distortion, so that's pretty dang high, as long as you aren't hitting it with blasts of air.
 
I think what Dave is saying is that saturation is an electronic function and is not what we are describing. A different process than what we are talking about. The changes in sound are just differences comparing in to out, so if you apply changes in pressure to a transducer that converts this change to changes in voltage, you should see a straight line on a graph. Dynamics (including ribbons) respond differently as the diaphragm/ribbon travels further. There will be distortion to the line when it slows down and goes back the other day. Mass, inertia and maybe friction will change the linearity. Condenser membranes are the lightest and should exhibit the best linearity. This is more a physics lab experiment. I’ve often wondered if again, it’s a less than ideal response we get that we like. It’s easy to wonder if kick drum mics sound good or bad because the diaphragm doesn’t do the job perfectly, but we like it! Has anyone ever tried to use a measurement omni for recording. Designed for the measurement task they sound really boring and unspecial.
 
"I think what Dave is saying is that saturation is an electronic function and is not what we are describing." Indeed Rob and I will admit I am being very pedantic. Certain 'technical' terms have entered the audio world and been debased and used wrongly "rms watts is another one but I freely admit, 'rms' and 'saturation' are here to stay and there is little we can do now.

There is an effect of saturation that I have observed several times in my career as a R&TV service man. In the event of a lighting strike, devices such as VCRs often escaped serious damage and needed little more than a fuse replaced, sometimes a mains filter cap. This was because I am sure, such kit USED to use a 50Hz mains transformer. The very high current saturated the transformer and it 'looked' like a near short circuit thus blowing the fuse and protecting downstream electronics. The TVs on the other had, which never had transformers was often a write off even though connected to the same mains outlet as the VCR. So called "surge protectors" were blown to bits, only the wires remaining.

Dave.
 
Interesting points. I'm learning a lot here.

I know that the starting point on a guitar amp for many is to balance the gain on the breakup point so it sounds clean when you play lightly but you can dig in to your pickups for the crunch. It's just a best of both worlds technique. You can lightly play chords without it turning into high gain mess. Or strum out powerchords for crunch. lightly play a lead part. I dunno, I tried to record that way once and was very dissapointed with how many balls my production lost. I'm gaining up to 7 or 10 most of the time. Those powerchords/chugs feel aenemic on any other setting.

The noise levels were really that bad in those old pre-amps in my first interface. Much more than ambient room noise. And to add to the fact that I needed to turn it up to max just to get a semi decent level when using the 57. I'm talking about the power of the pre-amps mostly, not the fidelity as such, noise levels are a non issue now, the 18i20 is perfectly useable. That apogee duet would run anything though. my focusrite 18i20 does not. it just about has enough guts to record with the 57 for most applications. But if I want to record a softer classical guitar part with my ribbon mic...pfft.. forget it.
Not unless I use cloudlifters or external pres, or just stick with condensers but.... why would I want to do that?

Maybe I need to invest in a decent workhorse condenser like the AKG414 or something. then perhaps I might actually start to appreciate the condenser mic.
 
Ha! I'm the reverse. I use the 414 for my go to mic, and have just ordered an inline preamp so I can see if I can perk up the sm7bI just treat as 'ok'?

Seriously though - my old Ferroraph reel to reel with 1/4" unbalanced inputs connected to my 57 in the late 70's had no issues with hiss - the thing that annoyed me on all analgue recorders. Once digital arrived I don't think hiss reared it's head again, apart from on dirt cheap dreadful devices!
 
paulears said:
Ha! I'm the reverse. I use the 414 for my go to mic, and have just ordered an inline preamp so I can see if I can perk up the sm7bI just treat as 'ok'?

Seriously though - my old Ferroraph reel to reel with 1/4" unbalanced inputs connected to my 57 in the late 70's had no issues with hiss - the thing that annoyed me on all analgue recorders. Once digital arrived I don't think hiss reared it's head again, apart from on dirt cheap dreadful devices!
Click to expand...
I would guess that 1970s 57 was a high impedance model and as such the internal transformer gave you a noise free 20dB boost. Tape machines of that day were also pretty noisy, likely no better than 55dB below 3%thd and so bit o pre amp noise was not noticed. Shoot! the machines themselves were pretty noisy! Why 'serious' studios had a 'tape room'.

Dave.
 
This thread is what happens when you mix personal perception, personal recording philosophy, technical specifications and anecdotal experience, then try to explain it in technical terms without necessarily having the technical background to do so.

The direct answer to the question is everything you change makes a difference. Whether those differences get in the way of creating a good recording is the question that should be asked. The answer is no.
 
Had and have the UX2 and its USB power supply is weak and the whole unit is weak in volume, but it can work really well for say one singer over some fruityloop beats program and a mic booster takes a dynamic to where the UX2 works fine in some applications.
The ADDA isnt the issue at all. The headphone amp is weak but with low ohm headphones or powered speakers, or even adding a rack mount headphone amp it works fine and is super cheap...and its main thing is the PODFARM software and portability.
Not saying its great, its actually bottom of the ladder level stuff and plastic that works....but the expensive stuff isnt a million miles better in completed sound either. Im amazed at some of the recordings off that UX2 with the F-loops stuff.

Upgraded to the UX8 and it was nothing like UX2 and was used on commercially released albums Todd Rundgren and Pyschedlic Furs and maybe others...2009ish piece and was a failed and obsoleted life span in competition to protools and other gear apparently. Super low latency still as good as others today claim. It was a big upgrade.

(However the other day one channel sounded like ass, then one by one I plugged in and only 3 were normal sounding of the 8!
I already new the Inst INputs fail with some odd capcitor issue a known old unit issue/failure and now the Mic/Line I/O are failing...with 3/8 working and really 1 of the 3 sounds "full".....so its time to trade it in. A lot of it works great, the headphone amp is really nice and the SPDIF works perfect....$50?$100? for old junk? I would highly recommend NOT buying one of theseUX8 at this age.)


2021....
I agree with those voting on buying for the I/O and SPDIF or ADAT or whatever functions one needs.
Like Dave and son, I tried a bunch of different interfaces before and didnt hear any huge sound convertor difference between Frite, Tascam, MOTU, EMU, others.... its small compared to a room noise or mic self noise. Is a Clarett vs a Scarlett the more modern question than UX2/UX8, probably. REVERB has the Scarletts as the biggest sellers.

Are new converters better than the old convertors? Old convertors used on platinum albums in the past sound good to me?
My inability to get a Billboard # 1 isnt the convertors fault. lol

My son needs more channels for drums and band album, so 8ch rack mounts might be the thing.
Behringer UM1820 $279 is probably what we'll get and then add the AD8200 for $219,,,,have 16 channel. new with warranty.. If it can last 5yrs Id be happy. Hes already got the band-multi headphone Behringer rack that has a couple bad channels these days and is >5yrs old. It was $50 used though too. 0% interest make some payments...16 channel mic pre>studio DAW > multi headphone amps and speakers! ~$600 not bad!

For me dabbling in very little recording these days but enjoying a hi end headphone amps and maybe interested in 600 ohm headphones a 2channel is fine..but I might appreciate the quality build more and need a SPDIF to feed the Grace Design HeadAMp I have and also accept the SPDIF IN from my Frite 430 vintage outboard with huge pumpkin transformers.... :)
 
Converters are really a non-issue. Metallica's Black album was mixed down to a 16 bit Sony dat machine using its internal converters 30 years ago. The album sold North of 20 million copies...
 
You must log in or register to reply here.
Back
Top