Audio interface preamp "sound" question

[alfaholiq]

Hello everyone,

It is often heard how some expensive interface preamps like Apollo sound better than, let's say Focusrite Scarlett for example. And this is somewhat natural, the price difference is there, both belong to different price segments, so I am not saying that this is not true, I do not know. Maybe it is.

However whenever I ask why is that, and what makes the difference if we know that those preamps are 'clean', so no harmonic distortion all the way down to -90db or more, the frequency curve is practically identical, so what else can give a difference in sound quality, or better say what else can give a different sound character, well then all the hell breaks loose and people start yelling and raging about how stupid the question is.

So let's try again, hoping that this community is less religious and more scientific, what can make the difference other than frequency response and harmonic distortion in an audio interface preamp?
Is there anything that can make the transients different?
Are there some 'magical' things that can not be measured?

 

[fpitas]

Is there anything that can make the transients different?
Are there some 'magical' things that can not be measured?

No, and no. Transient response is defined by frequency response.

 

[fpitas]

No offense, but we get this sort of question frequently. We've been measuring audio equipment for over 100 years now. The mysteries have all fled.

 

[fpitas]

You may find your answers here: https://www.audiosciencereview.com/...e-as-transparent-are-that-many-confused.9245/

 

[alfaholiq]

Thanks. The thing is that I already knew this, however so many people said they hear some difference, so I started thinking maybe I am crazy here and came here to ask the question.

 

[solderdude]

what can make the difference other than frequency response and harmonic distortion in an audio interface preamp?

Output level differences on direct comparisons.
Used reconstruction filters.

Is there anything that can make the transients different?

It depends on what you mean with transients. When you mean the shape of the waveform of the highest frequencies then yes, there are measurable differences depending on the filter type.

When you mean transients in music (it consists of transients only) then no because 'sharpness' of sound is in the 5-12kHz region and all DACs perform very much the same.
Only with a severely rolled off treble and young ears you could possibly detect differences.
This is because of filtering in this case.

Are there some 'magical' things that can not be measured?

Everything can be measured but not all IS measured.


Most of the perceived differences usually are not caused by technical issues/differences.

 

[fpitas]

Thanks. The thing is that I already knew this, however so many people said they hear some difference, so I started thinking maybe I am crazy here and came here to ask the question.

It's very common that people hear differences, differences that vanish when blind testing is performed.

 

[fpitas]

Many manufacturers depend on the fallibility of sighted hearing. Adding cost to a unit often makes for better reviews.

 

[DVDdoug]

Right, there is nothing "mysterious". *

All preamps generate noise (hiss) and of course some are worse than others. There are different ways of measuring & specifying noise so you can't reliably compare the published specs. USB power is notoriously noisy and with USB powered interfaces, sometimes that noise gets into the preamp (usually a high-pitch whine).

Some "classic" preamps, especially tube preamps, may have slight "desirable" distortion and frequency response variations. But there is not "one tube sound" and if a tube amp sounds different it will be different from every other preamp.

Thanks. The thing is that I already knew this, however so many people said they hear some difference, so I started thinking maybe I am crazy here and came here to ask the question.

The "audiophile community" is MOSTLY crazy. ASR and HydrogenAudio are rare exceptions. There is a fair amount of nonsense in the pro audio community too but it's not quite as bad and they don't get obsessed by cables and things like that.

See Audiophoolery.

* P.S.
There are some mysterious and/or hard/impossible to measure/quantify things in audio, but not with the electronics - For example, the "soundstage" illusion is an illusion and it depends on the listener, the recording, speakers, speaker position, room acoustics, or the headphones, etc.

People talk about headphone soundstage a lot. I never hear a realistic soundstage with headphones and I thought I was an exception until someone posted this headphone soundstage survey.

And with lossy compression there can be temporal/transient artifacts that don't show-up in traditional continuous measurements. Lossy audio is best evaluated with blind listening tests.

 

[Ricardus]

The opamps used in modern preamps have had slew rates (the ability for an amp to add voltage gain over time) faster than any musical transients for decades and decades. So if by transient response you mean the quick rise time of a snare drum, the analog part of the signal chain has been up to the task longer than I have been alive.

 

[alfaholiq]

It depends on what you mean with transients.

A transient is the initial peak of a sound, the first spike in the waveform. I can be measured easily with the compressors, the same as its release timing and shape, however I wondered whether different sound card integrated preamps can sound different because of that, because this will not show in the frequency curve, nor in the harmonic distortion tests.

 

[Deleted member 21219]

, well then all the hell breaks loose and people start yelling and raging about how stupid the question is.

That behavior is generally an indication that they don't know the answer or that there is no answer.

however so many people said they hear some difference, so I started thinking maybe I am crazy here and came here to ask the question.

In a world suffused with insanity, a sane person is deemed to be crazy. Glad you've found our little haven here. It has respite from that.

BTW .....I almost forgot .....Welcome to ASR!

Jim

 

[alfaholiq]

BTW .....I almost forgot .....Welcome to ASR!

Jim

Thanks.

 

[Rja4000]

what can make the difference other than frequency response and harmonic distortion in an audio interface preamp?

That's a very good question.
Actually that was the very reason I started measuring gear.

There are objective (measurable) differences between interfaces.

1. Some preamps use transformers
Transformers are bringing their share of issues: distortion varying widely with frequency and level.

2. If noise makes little difference nowadays, dynamic range varies between preamps. Mostly at high level (condenser mic driven hard, or line level signal).
If that makes a difference ?
Maybe. Depending on your recording constraints.

3. Preamps are fed by microphones, which are transducers. Most transducers are far from ideal. They may vary their frequency response with preamp input impedance, as an example. (Some preamps allow you to select between several impedances).
It's not the preamp frequency response that may vary, it's more likely to be the microphone's.

4. Low-end interfaces have limitations
Some will saturate before Full scale (I own a Motu 828 mk III which behaves like that, as an example).
Lower cost means more compromises.

5. Analog gear are more prone to degradation with time.

...

Are there audible differences in relation with those topics ?
If there are, they are mostly subtile.
But if they exist, be assured that they can be measured easily.

 

[fpitas]

be assured that they can be measured easily

Exactly

 

[alfaholiq]

Thanks. I am aware of audiophiles being audiofools very often with their set of myths, however I am a musician and audio mixing engineer so I often encounter my colleagues's funny beliefs, which are not the same but nonsensical in the same manner.

So it is clear to me what a transformer does to the signal, tubes, or solid state saturation, but my question was just about the audio interfaces which never use tubes or transformers and usually keep the distortion levels very low.

The thing is, I could swear that the old Digidesign 002 had some terrible preamps, however I never measured it to see what was happening actually.
In those days we all believed how Logic and Pro tools render the audio better than the old Cubase VST32, I personally tested it then and heard how broken it was, until a year ago I installed it on some old machine and found out that it renders the audio normally, just like any other piece of software.
So, who knows what was that all about the Digi 002.

 

[AnalogSteph]

It is not uncommon for mic preamps of the simpler kind to run rather low on GBW near maximum gain... I bet a THD sweep / multitone / CCIF IMD / SMPTE IMD measurement would be revealing. The Behringer UMC20xHD and similar preamp have quite well-documented high H2 levels, for example. (Some measurements were posted on ASR fairly recently but I can't seem to find them for the life of me. Anyone else have better luck?) In lieu of the Digi 002, I might suggest an Arturia Minifuse 1 as an inexpensive "guinea pig" - people don't seem to like the preamps on this series very much.

Apparently there really was a DAW with rather broken processing and resulting "sound" at some point, though I don't remember what it was or which version. The problem may not manifest itself audibly right away but rather require some processing to occur for rounding errors to accumulate.

 

[solderdude]

A transient is the initial peak of a sound, the first spike in the waveform. I can be measured easily with the compressors, the same as its release timing and shape, however I wondered whether different sound card integrated preamps can sound different because of that, because this will not show in the frequency curve, nor in the harmonic distortion tests.

Ah... so you mean transients in music and not as in dV/dt rise/fall times of electronics.

Just zoom in on those transients and you will see they aren't nearly as 'fast' (dV/dt) as any electronics can handle (full power/voltage bandwidth) and usually span several samples even in 44.1kHz sampled files.
So any amplifier will just 'follow' or 'amplify' that waveform. It is perfectly captured with steady-state signals.
Now... when you are talking about transducers and acoustics... that's an entirely different ballgame.

 

[MaxwellsEq]

However whenever I ask why is that, and what makes the difference if we know that those preamps are 'clean', so no harmonic distortion all the way down to -90db or more, the frequency curve is practically identical, so what else can give a difference in sound quality, or better say what else can give a different sound character, well then all the hell breaks loose and people start yelling and raging about how stupid the question is.

When two objects are compared, it's imperative that the levels are very accurately matched between them.

If there is even a slight difference in level (so small that you can't hear the difference in volume), the slightly higher output device will generally sound "better". This is a major reason why people express a preference between two devices, with measurably good enough characteristics (amplitude response, noise, distortion).