D/A A/D Converter. Essential?

[Dean7]

Originally Posted by TurboJets

I had the converters in my 1814FW upgraded and the difference was more than noticeable, it was remarkable and notable. Like the proverbial "blanket being lifted" from my tracks. I was already getting to a point where I didn't have to EQ many of my tracks during mixdown and now my need for EQ is even less because the increase in clarity helps me hear more of the subtle differences between my mics and pre's. The change has also allowed me to hear more of my room's characteristics when tracking, mic proximity effects, tube characteristics in my pre's, etc. More than ever I'm able to discern the right pre and right mic for the job.

"Hearing" is everything, so it simply stands to reason that if you're hearing more clarity and accuracy then you're going to have a greater opportunity to become a better engineer.

Having said that I, will also say that I personally owe Ethan many thanks for his published material and the advice he imparts on forums like this. I've learned more than I can say from Ethan's advices.

Just my 2 cents.

Did you have JUST the converters upgraded? Or was it converter/pre/clock/other? Was it the black lion mod?

[TurboJets]

Yes, it's the Black Lion mod with power supply, analog path and converters upgraded.

[cowrange]

Ok first off, I agree that quality goes down the chain, in general. Player->Room->Instrument->Mics->Pres->Converters. I think anybody agrees that if you have a crap sound, good gear just picks up the crap more clearly. That being said, I've been using a 002 for a long time and was always frustrated at the smeary haze that the 002 converters put on stuff. I borrowed a friend's mytek and I finally heard what I had been missing all along. I show most guys that enter my studio an A/B test where I switch between the mix going out the 002 and then the mytek. They are matched level exactly. Everybody I've shown is amazed at the difference. Most of them are uneducated ears. Can I make a record without em? Hell yeah. Do I get much more satisfaction with Mytek? Amen. This bantering is beyond me. Yes already, if your mics are MXL, don't spend your next bills on a crazy good converter. This is common sense. But to say you need scientific data. Hmmm. I use my ears. Yes to me it's a huge difference. I'm now the proud owner of a Mytek 8x192. And yes I have great mics. Scheops, coles, aea, neumann, soundelux etc. I'm sure a crate amp miced with an MXL into some behringer gear will still sound crap through a Mytek. I bought great instruments first, then mics, pres and finally converters. Ok my rant is done. Peace all.

[Ethan Winer]

Originally Posted by nickynicknick

I really don't need science (although science is extremely helpful) to prove what is good and what is not

Just to be clear, I'm not here to start trouble or pick fights. And we do need science to separate fact from fiction. Time and again I've seen people express beliefs that, when put to the test, turn out not to be true. I've experienced this enough times personally to know that hearing and perception are very fragile.

The last time this happened was only a few months ago, after I burned a CD of a mix. I put the CD in my computer's CD drive and it sounded soft and muffled compared to the Wave file source I had played only five minutes earlier. I'm thinking WTF, why does this sound so different? So I extracted the Wave file from the CD, and lined up both Waves files in a DAW. Then they sounded identical. The problem was simply that the CD playing software was much softer. It sounded muffled! But it really wasn't different at all from the source file.

This is why level matching is critically important when doing A/B tests. So often I see people say that switching to this mic pre or that A/D/A converter was like lifting the proverbial veil. But when compared properly the differences disappear. I'm not saying that all gear sounds the same! But without a "scientific" comparison you can't know for sure.

--Ethan

[Slap Back]

Originally Posted by Ethan Winer

Just to be clear, I'm not here to start trouble or pick fights. And we do need science to separate fact from fiction. Time and again I've seen people express beliefs that, when put to the test, turn out not to be true. I've experienced this enough times personally to know that hearing and perception are very fragile.

The last time this happened was only a few months ago, after I burned a CD of a mix. I put the CD in my computer's CD drive and it sounded soft and muffled compared to the Wave file source I had played only five minutes earlier. I'm thinking WTF, why does this sound so different? So I extracted the Wave file from the CD, and lined up both Waves files in a DAW. Then they sounded identical. The problem was simply that the CD playing software was much softer. It sounded muffled! But it really wasn't different at all from the source file.

This is why level matching is critically important when doing A/B tests. So often I see people say that switching to this mic pre or that A/D/A converter was like lifting the proverbial veil. But when compared properly the differences disappear. I'm not saying that all gear sounds the same! But without a "scientific" comparison you can't know for sure.

--Ethan

In my experience, adding a single piece of gear and using it for an extended period of time, say 2 or 3 weeks or longer, will most accurately show the difference it makes. This way more factors get considered from many different angles.

[TurboJets]

So, you can "see" improved clarity in a waveform? I didn't realize you should be able to see more transients or see less mud in your captures. That would be pretty cool.

All I know is I've been home-recording for 13 years now and while I know that's not as long as a lot of other's here, I feel pretty confident in my ability to comment accurately on the gear I have experience with - the gear I've used. I tracked with the M-Audio 1814FW pretty heavily for a full year, had it modded by BLA and the mod made a huge difference in what I hear now in my captures and playback using the same mics and pre's and instruments.

When I listened to playback of projects I had already recorded through the 1814 [pre-modification] I was able to hear distortion, smearing, brittleness (etc.,) that I couldn't hear before but was definitely making me unhappy with my captures and mixes. I am getting sounds now that make me very happy and I'm so much more satisfied with how my recordings sound, and all I changed was the converters and analog path in my interface.

Scientific evidence is nice, even necessary in some cases. In this case, not so much. I don't need to see scientific evidence to prove to me that upgraded converters make a difference in your recordings. I'm sure the data exists somewhere to prove the case but when you hear the difference yourself that's usually all the evidence you need to bring a smile to your face. But that's just me.

[Dean7]

Originally Posted by TurboJets

Yes, it's the Black Lion mod with power supply, analog path and converters upgraded.

Well, that's upgrading a lot more than the conversion. I do agree that there are a TON of things you can do in your signal path to make things sound better. However, the actual conversion of the audio from analog to digital doesn't make as much difference as most people think. If you could arrange a test where the analog circuitry and signal was exactly the same right up to the ADC I bet most people would be amazed at what little difference it makes.

I'm not saying that you can't hear a difference. I just upgraded from an MBox 2 to a ProFire 2626 and there is a huge difference, but it's mostly in the mic pres and circuitry leading up to the ADC, not the conversion itself. It's just really freaking hard to figure out what difference you're hearing (e.g. is it a difference in the pres/converters/clock/whatever). So, that's why I am in favor of scientific tests mixed with common sense so that we can figure out WHY some things sound better than others.

-Dean

[Dean7]

Originally Posted by Ethan Winer

... And we do need science to separate fact from fiction. Time and again I've seen people express beliefs that, when put to the test, turn out not to be true. I've experienced this enough times personally to know that hearing and perception are very fragile....

Yeah, I've experienced that myself. I can buy a new mic or something and think it's amazing, but then I'll listen to it back to back with the mic it was replacing a couple months later and discover that it sounds almost identical, maybe worse. Mics are more of a subjective thing, so I don't think scientific tests really work there (since the flattest mic in the world might not be the best-sounding mic in the world), but I was just trying to illustrate the fact that sometimes our ears can deceive us. I'm not sure if it's because when I buy new gear or listen to new gear I really want it to sound better (because who doesn't want some magical piece of gear that will make their recordings sound better?) or what, but this has happened to me so much I kind of feel like I have to own something for at least a year to really know what it sounds like.

[reallyjake]

Originally Posted by Ethan Winer

The only things that matter with audio are frequency response, noise, distortion, and time-based errors. Of course there are subsets - hum and crackling both fall under noise - but those four parameters define everything that matters with audio.

If you look at actual components and how they spec, often you find there might be a little more to it than those four parameters. Tonality is hard to measure for instance. I guess you could "try" to lump that in with frequency response, but it would be misleading based on how most measure it.

So tonality, it can be ambient, warm, sweet, clinical, sterile, etc. One company can use parts whose sum total equal the specs on paper of another company's product. But the sum total of those parts may have a different tonality.

Also, consider this: Teflon capacitors have just about the best spec of any capacitor. Degree of losses for frequencies are almost non-existent. But if you have a poor circuit design, it might be best if some of those frequencies NOT be heard.

So one company's converter design might achieve a pleasant tonality out of clever design, and another might chop and block until the spec looks the same on paper. But they won't sound the same.

[reallyjake]

And I'm still happily using my Lynxone, which is over ten years old , right?

So I'm by no means saying go out and buy the most expensive thing. Converters hit a diminishing return. I keep seeing people go from low end converters to buying the most expensive "new thing" and I'm not sure that dramatic of a step is needed.

[Ethan Winer]

Originally Posted by Dean7

I can buy a new mic or something and think it's amazing, but then I'll listen to it back to back with the mic it was replacing a couple months later and discover that it sounds almost identical, maybe worse.

Exactly. And before you did that side-by-side you were just certain your new A/D converter - oops, I mean microphone - was much better than the last one. This happens all the time, yet people refuse to understand how short term auditory memory really is. Just as important, a valid comparison must test the same source material. Every session is different, even if it's the same band. Move a microphone half an inch and mud can turn to sparkle or vice versa.

Mics are more of a subjective thing, so I don't think scientific tests really work there

I assure you that "science" applies to microphones too.

(since the flattest mic in the world might not be the best-sounding mic in the world)

Well, let's examine that in more detail. What can vary between microphones? Aside from residual noise and distortion, the main difference is frequency response, both on and off axis. Let's forget off-axis response for the moment and assume a single instrument or singer being recorded with one microphone on-axis. And no other instruments are playing in the room where the off-axis response would matter. And we'll also ignore susceptibility to popping and sibilance. I'm talking purely the microphone's sound quality.

So now the only significant difference between any two microphones is frequency response. Of course, this can vary with distance (proximity effect), so let's further assume the musician does not move to ensure the LF response stays constant.

If you compare one microphone that's flat, with another microphone that has a presence peak or whatever that falls at a pleasing frequency, there's no reason simple EQ cannot make the flat microphone have that same pleasing response. Of course, some microphones have no output at all at very high frequencies, so no amount of EQ can bring that content back. But other than that, I see no reason a flat microphone can't be EQ'd to sound like another microphone that has a response you like.

I'm not sure if it's because when I buy new gear or listen to new gear I really want it to sound better

Yeah, that'll do it every time. The more you spent, the better it sounds. Guaranteed.

--Ethan

[Ethan Winer]

Originally Posted by reallyjake

Tonality is hard to measure for instance. I guess you could "try" to lump that in with frequency response, but it would be misleading based on how most measure it.

It's not misleading! Tonality is by definition frequency response (including harmonic content added by distortion). What else could it possibly be?

One company can use parts whose sum total equal the specs on paper of another company's product. But the sum total of those parts may have a different tonality.

It doesn't work that way. There's no magic to this. Really. And nobody designs gear to "chop out" frequencies to match somebody else's gear or to match some perceived level of quality.

--Ethan

[nickynicknick]

So there it is....the OP is gone because Ethan and his buddy want to pat each other on the back....nice job guys...your opinion is old now...we get it...we're not allowed to have an opinion based in experience. According to you two superior minded engineers none of our opinions matter unless we can prove it empirically. I was kind of interested in the OP, but now I can't even remember what that was...but I would like to say...IMO good conversion is an absolutely essential part of your recording set up and you definitely should be able to hear the difference between a motu (OK conversion) , a Rosetta (pretty good conversion)and a Prism(really good conversion)...now which of these works best in any given situation is very debatable...high end isn't always better...but that's a different story...now Ethan, unless you can provide a null test that proves that you,in fact, actually exist.. please don't respond to this as this an opinion based completely in my personal experience which, according to you is unimportant.

Nick

[Dean7]

Originally Posted by Ethan Winer

Exactly. And before you did that side-by-side you were just certain your new A/D converter - oops, I mean microphone - was much better than the last one. This happens all the time, yet people refuse to understand how short term auditory memory really is. Just as important, a valid comparison must test the same source material. Every session is different, even if it's the same band. Move a microphone half an inch and mud can turn to sparkle or vice versa.

Yeah. I've done tests with A/D converters as well that have yielded the same results. I'm not sure what the "answer" is in terms of my quest for great sound, but I've decided that I can't trust anybody, because most of the guys giving advice are giving that advice based on the fact that they just paid 5K for something and they really don't want it to not be the cat's meow.

Originally Posted by Ethan Winer

Well, let's examine that in more detail. What can vary between microphones? Aside from residual noise and distortion, the main difference is frequency response, both on and off axis. Let's forget off-axis response for the moment and assume a single instrument or singer being recorded with one microphone on-axis. And no other instruments are playing in the room where the off-axis response would matter. And we'll also ignore susceptibility to popping and sibilance. I'm talking purely the microphone's sound quality.

So now the only significant difference between any two microphones is frequency response. Of course, this can vary with distance (proximity effect), so let's further assume the musician does not move to ensure the LF response stays constant.

If you compare one microphone that's flat, with another microphone that has a presence peak or whatever that falls at a pleasing frequency, there's no reason simple EQ cannot make the flat microphone have that same pleasing response. Of course, some microphones have no output at all at very high frequencies, so no amount of EQ can bring that content back. But other than that, I see no reason a flat microphone can't be EQ'd to sound like another microphone that has a response you like.

But, I was talking about strange trashy-sounding mics that just sound "bad", but sometimes "bad" is what you were looking for (sometimes I experiment with a really trashy room mic that I can add in to the mix at ultra low volumes to add some texture, for example). In terms of a good mic that reproduces what you hear in the room, I agree with you.

Originally Posted by Ethan Winer

Yeah, that'll do it every time. The more you spent, the better it sounds. Guaranteed.

Yep. I try to guard against it these days by assuming that everything is going to sound horrible. :)

-Dean

[Ethan Winer]

Originally Posted by Dean7

But, I was talking about strange trashy-sounding mics that just sound "bad"

Okay, I don't usually even think about gear like that.

Why do people get so damn hostile about this stuff?

--Ethan

[kedbear]

Originally Posted by Ethan Winer

Exactly. And before you did that side-by-side you were just certain your new A/D converter - oops, I mean microphone - was much better than the last one. This happens all the time, yet people refuse to understand how short term auditory memory really is. Just as important, a valid comparison must test the same source material. Every session is different, even if it's the same band. Move a microphone half an inch and mud can turn to sparkle or vice versa.



I assure you that "science" applies to microphones too.



Well, let's examine that in more detail. What can vary between microphones? Aside from residual noise and distortion, the main difference is frequency response, both on and off axis. Let's forget off-axis response for the moment and assume a single instrument or singer being recorded with one microphone on-axis. And no other instruments are playing in the room where the off-axis response would matter. And we'll also ignore susceptibility to popping and sibilance. I'm talking purely the microphone's sound quality.

So now the only significant difference between any two microphones is frequency response. Of course, this can vary with distance (proximity effect), so let's further assume the musician does not move to ensure the LF response stays constant.

If you compare one microphone that's flat, with another microphone that has a presence peak or whatever that falls at a pleasing frequency, there's no reason simple EQ cannot make the flat microphone have that same pleasing response. Of course, some microphones have no output at all at very high frequencies, so no amount of EQ can bring that content back. But other than that, I see no reason a flat microphone can't be EQ'd to sound like another microphone that has a response you like.



Yeah, that'll do it every time. The more you spent, the better it sounds. Guaranteed.

--Ethan

But, like, could this not just be said about a whole host of things. I mean, this may sound stupid, but if you made some cheap bright grand piano have the same tonal/frequency response as an expensive bechstein grand, are they going to sound the same?

Generally, eq'ing bright guitar strings doesn't sound as good as using softer strings (i don't like elixir's), or eq'ing a bright cheap mic does not sound as nice as starting with a good mic with the frequency response you were going for.

[reallyjake]

Originally Posted by Ethan Winer

It's not misleading! Tonality is by definition frequency response (including harmonic content added by distortion). What else could it possibly be?

Look, it is misleading. Find me two converter companies' spec sheets who you can reliably compare tonality. You can compare sample rates all day long. But not the idiosyncrasies of tone. You also will not find specs of the degree of loss in any given frequency range. But again, you can compare sample rates all day long. So yes, it is misleading to pretend we can get everything handed to us on a data sheet. It is also misleading to pretend we can easily measure tangental information in the average studio or just find this info easy just using a few household items.

But most of us can use our ears.

Sound and sonics are not an easy thing. Just because your definition is easy, doesn't make frequency analysis easy. You cannot pretend that a response is only a few things that can easily be measured over their entire spectrum and posted on a company website. And that such a misguided attempt would spell out the entire picture of a product's sound.

That's just bizarre to even consider a tonality comparison on a data sheet. Especially for something like converters. There are too many variables. Harmonic distortion varies so much, these companies do not even attempt to measure it over the whole frequency range. You can say it's here or there. But it won't be in a data sheet. Do you really think a company is going to measure even order harmonics? Losses or perceived boosting over an entire frequency range? Again, maybe you could measure this or that over 100 Hz, 1 KHz, and 10 KHz. But that's only a snapshot in the family album.

Do you think all distortion characteristics are the same? And it's variable after variable.

So yes, it is misleading to clump all these things under frequency response. It's even more misleading to say this is all spelled out, we just need to look at numbers, etc.

[TurboJets]

Originally Posted by reallyjake

Do you think all distortion characteristics are the same? And it's variable after variable.

So yes, it is misleading to clump all these things under frequency response. It's even more misleading to say this is all spelled out, we just need to look at numbers, etc.

This is what I agree with most. Distortion characteristics are not the same. You can listen to 2 tracks that measure the same amount of distortion in a specified freq. range all day but that distortion will have different characteristics inciting different human descriptions. Sometimes vastly different descriptions.

It's the reason people use adjectives instead of numbers and measurements to describe the virtues and/or shortcomings of microphones, mic pre's, converters that are part of your interface or signal chain, and other outboard gear. Adjectives are discussed and given far more consideration than the specs. Specs are part of the decision making process yes, but the purchase is usually governed by what we hear when auditioning gear and by what we hear when people post samples of their private tracks and A/B tests when we're sharing the excitement of a new addition to our studio's.

Most of the A/B tests we hear on the internet are scientifically inaccurate - true. But the unique human experience of listening and hearing is what audio engineering is all about - we're not stupid. Without the human factor all this gear would just be worthless piles of metal and components, right?

The numbers and science are great guidelines that everyone considers - for sure. Tonality, color, and character is what we end up buying.

[Aaron Miller]

I'd love to hear what Dan Lavry has to say about all this!

[Dean7]

Originally Posted by Ethan Winer

Okay, I don't usually even think about gear like that.

Why do people get so damn hostile about this stuff?

--Ethan

Hah. Well, I make my recordings in my brother's untreated basement, so we have to get pretty creative. :)

I think people get hostile about it for the following reasons:

1. You're telling them that the 5k piece of gear they just bought makes .004% difference in quality, and people don't like to hear that. A lot of times, when people buy something they convince themselves that it's the cat's meow no matter what to prevent buyer's remorse.

2. You're telling them that they need to do the not-so-fun task of room treatments and learning how to use their gear if they want to make better recordings instead of just collecting more gear (which we ALL love to do).

3. Whenever anybody presents an idea that's contrary to what people have heard for the past x years they tend to fight it (no matter what it is) initially, and acceptance will come later. :)

-Dean

[Ethan Winer]

Originally Posted by kedbear

this may sound stupid, but if you made some cheap bright grand piano have the same tonal/frequency response as an expensive bechstein grand, are they going to sound the same?

Not stupid at all. But there's more to what makes pianos sound different than simple frequency response. Each note has a tonal balance between the fundamental and all of its harmonics. But each note also decays differently based on the piano's self-resonances. So using EQ to make the frequencies in an A chord on one piano match the same A chord on another piano may not make both pianos sound the same when actually played. Further, pianos respond to loudness differently, so any matching would also be valid only for that one volume level.

My point in all this is there are no "unknowns" in the science of audio. If two things sound different, that difference can be identified. Whether you can use EQ to match things is a different issue, but to my way of thinking there's no reason two microphones can't be EQ'd to sound the same, or at least very close. Versus pianos which are far more complex.

The limitations with EQ are as I described earlier - if a microphone cuts off sharply at either frequency extreme it's impossible to bring out those missing frequencies. Another issue I didn't mention is microphone self-resonance. EQ can add a resonance to a flat microphone to match it to another having a resonance. And EQ can reduce the level of a resonance to match it to a flat microphone. But EQ can't remove the extended decay that always accompanies resonance, so that aspect cannot be matched using EQ. But I bet you could get pretty close if you're able to accurately identify the frequencies. This is difficult to do by ear, but could be accomplished easily using an FFT plug-in.

--Ethan

[Ethan Winer]

Originally Posted by reallyjake

You also will not find specs of the degree of loss in any given frequency range ... That's just bizarre to even consider a tonality comparison on a data sheet ... Harmonic distortion varies so much, these companies do not even attempt to measure it over the whole frequency range.

Originally Posted by TurboJets

This is what I agree with most. Distortion characteristics are not the same.

Since you're both asking about the same thing, I'll address that too.

First, tonality is frequency response - by definition. "Loss in any given frequency range" is exactly the same as frequency response. You understand this, yes? So why do you think that a device that measures flat within, say, 0.10 dB from 20 Hz to 20 KHz will even have a "tonality?"

As for different types of distortion sounding different, this is correct. Most "analog" type distortions have a similar spectrum, where the lower harmonics are most prominent, and the level falls off with rising frequency. And with digital there's also aliasing which is non-harmonic and thus more objectionable. But - and this is the big one - if the sum of all artifacts is sufficiently low to be inaudible, then the nature of the distortion doesn't matter because you can't hear it. This is why accepted spec practice these days is to lump noise and all distortions into a single figure, given as some number of dB down.

Again, masking is the key. If all artifacts are 110 dB or more below the signal, then the specific spectrum of the distortions is irrelevant.

--Ethan