D/A A/D Converter. Essential?

[Ethan Winer]

Originally Posted by synthoid

I think it's pretty misleading to suggest that low-end converters and high-end converters differ in quality by 1% or something of that kind.

Okay, by how much do they vary exactly? Please be very specific and complete in your explanation!

I dunno Ethan. I've seen you arguing that Mackie monitors are as good as Adams and now that differences in converters are largely insignificant. Makes me wonder whether you listen to gear or simply read frequency response graphs.

I play, record, mix, and listen to music all the time. Again, it depends on how you define better. To me, "better" means wider and flatter frequency response, lower ringing, lower distortion, greater s/n ratio (gear) or higher maximum SPL (speakers), and so forth. I'm not married to any piece of gear or brand. I know lots of people who are all satisfied with many different loudspeakers. All I care about is the science.

A perfect example is loudspeakers that aim to be flat, versus loudspeakers that aim to flatter the music and sound pleasant. Which do you prefer to mix on, and why?

--Ethan

[Bierce85]

Originally Posted by Ethan Winer

The ratio of direct to reflected sound isn't really the issue here. Rather, it's that nobody can sing the same way twice, nor avoid moving their head and angle at least a little. I imagine "s" and "z" sounds in particular will vary a lot from one take to another for both reasons.

I don't understand why it isn't. I think your test varies from this one as far as conditions go so it's not really all that scientifically sound to apply your findings in that test to this one. A singer 2 inches from a mic is different from measuring sine waves through loudspeakers. The same rules of acoustics apply to all sounds but this is still different.


As for varying performances, I understand the fact that no one can sing the same exact way twice but is it not the same voice on both files? Differences in performances don't change the overall tonality or clarity of the voice and that's what changed between the two files. "s" and "p" sounds maybe, but that's just one thing that changed between the files.

Just because two converters have the same specs does not mean they'll sound the same. I wish it did so I could just read a spec sheet and buy one based on that, but that's just not the way it goes.

[Ethan Winer]

Originally Posted by Bierce85

I don't understand why it isn't.

I don't know what else to say that I didn't already explain in great detail above.

I wish it did so I could just read a spec sheet and buy one based on that, but that's just not the way it goes.

What aspect of audio fidelity do you believe specs overlook?

--Ethan

[Bierce85]

you explained how the performance can differ, thus yielding a different sounding recording. I'm not refuting that. What you didn't explain is how the same results you got measuring sine waves through 2 loudspeakers picked up at the listening position (I would assume 3-5 feet away) can be applied to a singer 2-4 inches from a cardioid pattern microphone. It's quite simply a different scenario which I believe would not have nearly as varying results.

My opinion is that there is a clear difference in conversion quality in favor of the mytek, and the imperfections of the test are not enough to debunk them. Since it's not a perfect test, I can't be 100% sure the differences I'm hearing are the converters, but I'm 95% sure they are and that's enough for me.

This is of course, my opinion.. but I really do believe room acoustics are a negligible issue here.

[Ethan Winer]

Originally Posted by Bierce85

What you didn't explain is how the same results you got measuring sine waves through 2 loudspeakers picked up at the listening position (I would assume 3-5 feet away) can be applied to a singer 2-4 inches from a cardioid pattern microphone. It's quite simply a different scenario which I believe would not have nearly as varying results.

It's not so much distance as angle, and not being able to sing the same way twice. But distance is a factor too. If you're two inches from a microphone and move only half an inch, that's a change of 25 percent! Proximity effect alone will probably change the LF response by several dB with a 25 percent difference.

I really do believe room acoustics are a negligible issue here.

I agree, but it could be a factor. It depends on how close any reflecting surfaces are. In a totally dead room I agree, but there's still the angle and performance issues.

--Ethan

[Bierce85]

Originally Posted by Ethan Winer

What aspect of audio fidelity do you believe specs overlook?

--Ethan

I believe higher voltage discrete designs will sound better than inexpensive integrated ones. Whether or not that can be reflected completely by specs is beyond my knowledge. I know things like THD and frequency response tell you something but I don't think they tell you more than your ears can, or perhaps they can but really in a different, less important way. I'm no electrical engineer, but that's what I believe.

Originally Posted by Ethan Winer

I agree, but it could be a factor. It depends on how close any reflecting surfaces are. In a totally dead room I agree, but there's still the angle and performance issues.

--Ethan

I guess you're just a "needs total scientific proof" type of guy and I'm more apt to go with what seems the most likely to me. I'd still say that even if half of the difference between the two files are performance/angle/distance/room, the other half is still worth the money for the better converter.

[Ethan Winer]

Originally Posted by Bierce85

I believe higher voltage discrete designs will sound better than inexpensive integrated ones.

That may be your belief, but there's no basis in fact for believing that. Assuming you don't clip because of inadequate headroom. 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.

I know things like THD and frequency response tell you something but I don't think they tell you more than your ears can, or perhaps they can but really in a different, less important way.

Ears are a terrible test instrument! The same signal heard on different days can/does seem different, and ears have nowhere near the resolution and dynamic range of test gear. Have you never made a mix you thought was great, only to think it sucks the next day? I rest my case.

I'd still say that even if half of the difference between the two files are performance/angle/distance/room, the other half is still worth the money for the better converter.

Look at it this way - if the difference between two devices is "huge" as was reported, something measurable must be very different. Perhaps a 4 dB peak or dip in the response, or 3 percent distortion, or some such. As I'm sure I said earlier, it could be that one of the devices was outright broken. But if two devices spec well, and they both meet those specs, and one sounds "very different" than the other, I have to suspect the listening test procedure. Hence my objection to comparing two different performances.

If you have a better explanation, I'm all ears!

--Ethan

[Bierce85]

Ethan,

All your points are valid and I appreciate you taking the time to have this discussion. I'm glad there's people like you on this board who try to represent truth instead of salesmanship (or both mixed together when appropriate )

You may very well be right but I think it's an awful big coincidence that in this test the Mytek sounds better in all the ways you'd expect.. detail and clarity. I would be much more apt to believe the test caused the difference if it was the other way around and the Digi002 sounded clean while the mytek sounded veiled. Maybe that's me falling for converter marketing, maybe not.

I don't own high end converters and would love to believe there's no need for them but until someone can post a perfect shootout between some low end and high end converters in which there's no audible difference or the difference is negligible, my opinion wont change. Despite that, I don't plan on upgrading from my fireface for a long time. Preamps, mics and acoustic treatment all are a lot more important to me right now.

[craigmorris74]

Aw crap, I must have bought the wrong mic pres! The Mackie Onyx pres have MUCH better specs than my Neve and API modules!

[Bierce85]

Originally Posted by craigmorris74

Aw crap, I must have bought the wrong mic pres! The Mackie Onyx pres have MUCH better specs than my Neve and API modules!

I don't think this is really relevant.. Ethan never made any claims that better spec'd preamps are going to sound better to your ears.

[corruptmary]

Several people here seem to be blaming different performances for the differences in converter shootouts, but it's not always the case. The converter test I did at work was taking the same audio file and running it through different converters. There IS a difference.

However I do agree with the people who said that after a certain price point you get less value for the money you're spending.

[Ethan Winer]

Originally Posted by Bierce85

the Mytek sounds better in all the ways you'd expect.. detail and clarity.

I can make any device sound more detailed and open by simply boosting 10 KHz a dB or two. Likewise for warm - just shave a few dB off the extreme highs and maybe add some tube or transformer distortion.

Perception is everything, and nobody is immune from the influence of advertising, including me.

Preamps, mics and acoustic treatment all are a lot more important to me right now.

For sure.

Good thread man.

--Ethan

[Ethan Winer]

Originally Posted by corruptmary

There IS a difference.

I'm sure there can be differences! But when someone proclaims a "shootout" we need to know that what's being compared is correct. That's my only point here.

We also need to distinguish raw fidelity / clarity / transparency etc from "sounds good" as in tubes and transformers and intentional color.

--Ethan

[craigmorris74]

Originally Posted by Bierce85

I don't think this is really relevant.. Ethan never made any claims that better spec'd preamps are going to sound better to your ears.

OK, removing sarcasm, according to his other statements, Ethan would dig the Mackie Onyx more than an API 512c or a Neve 1072. Better specs. That's a fair statement.

I do think Ethan's wise in trying to get us to not fret so much over gear, especially converters. I'm quite certain (even though I wouldn't want to), I could make a great recording of a great band in a room that didn't sound offensive with an SM57and a cheap condenser ran into the mic pres and converters on an M-Audio box, as long as the room doesn't suck.

[primemeat64]

Originally Posted by Ethan Winer

It would be 0.4 percent better.

Seriously, outboard A/D/A is probably the last thing you need to worry about, and it comes way after having good microphones and good monitoring in a good room.

--Ethan

Ethan hit the nail on the head straight away... Arguably the most important thing is the quality of song, the musicians and their instruments - once into the recording side of things yeah, mic's, room and monitoring (and engineer skills!) will impact the mix a LOT more than your converters.... unless perhaps you're using the line-in of your PC's motherboard!

[mytek]

Quote:
Originally Posted by Bierce85
"the Mytek sounds better in all the ways you'd expect.. detail and clarity"

I can make any device sound more detailed and open by simply boosting 10 KHz a dB or two. Likewise for warm - just shave a few dB off the extreme highs and maybe add some tube or transformer distortion.

--Ethan

I'll chime in here , maybe you'll find it valuable. I think both you guys have a merit however it sounds a bit like comparing apples and oranges.

Room sound and digital electronics sound are completely different domains and while they overlap- (you can't do one right without doing both of them right) the aestetics of perception are different- different things get distorted in different ways.

In Mytek case, to design and test and A/B compare converters we had to built a properly acoustically designed flat sounding floated mastering mastering room, to get rid of a veil and confusion caused by bad room (and yes we measured the room to get that right). Then we had to get hi res monitoring system with hi-fi and good speakers (Lipinski in our case). Then only could we really hear what's going on when we change a chip in a converter or use different cap. (which by the way is often impossible to measure with standard instrument measurement).

Also bear in mind the most nasty digital distrotion are non harmonic ie not related to music material that's why you hear them even when they are at -100dB.

BTW. You can't turn m-audio into pacific microsonics by boosting 10k.

Regards, Michal , Mytek New York

[glowworm]

can we get a really good scientific shootout by you mytek guys then? I heard the difference. for my money I stepped up from the 002 to the 003 and could hear a big difference in that recording path. does anyone know a good dedicated DA converter for those of us who don't do allot of ADing at home and just need a good DA to listen to?

[Lunatique]

Originally Posted by glowworm

can we get a really good scientific shootout by you mytek guys then? I heard the difference. for my money I stepped up from the 002 to the 003 and could hear a big difference in that recording path. does anyone know a good dedicated DA converter for those of us who don't do allot of ADing at home and just need a good DA to listen to?

Yeah, I second that request. I think if premium A/D D/A converter makers all posted A/B tests of their converters against various popular audio interfaces, they'd be converting a bunch of new customers (no pun intended) immediately. Most of us have searched high and low for such A/B tests online, and to put it bluntly, we've all got one hand on our wallets ready to pull it out if we could just be convinced that the investment is worth it.

[atticus]

Here is a link to the sample page on our website. Thanks!!

MYTEK DIGITAL USA

[s0nguy]

Great converters will change your world completely. You wont be trying to figure out how to get "that sound"

I tested a ac guitar with a Neumann km184 through a Great River MP2nv to A Digi002 and it was like.. meh.. I added the UA 2192 using the same pre and mic and it was nothing less than a life changing moment...an "AH-HA!! so THAT"S how they do it." moment.

Once you start stacking tracks with great conversion the difference becomes even more embarrassing.

Whoever says its a minimal difference has never experienced a truly great converter/clock.

ymmv,
-s0nguy

[Ethan Winer]

Originally Posted by craigmorris74

according to his other statements, Ethan would dig the Mackie Onyx more than an API 512c or a Neve 1072. Better specs. That's a fair statement.

It seems to me the best way to record is to capture sounds as cleanly as possible. Then you can dirty them up in mixing as needed. Related, I'm a big proponent of deferring decisions and using non-destructive editing and effects. Maybe it's because I can never make up my mind if it sounds good until I've heard a mix many times. Better engineers than me can make decisions on the spot that hold up over time. Me, I fuss and fiddle endlessly.

--Ethan

[s0nguy]

Originally Posted by Ethan Winer

Maybe it's because I can never make up my mind if it sounds good until I've heard a mix many times. Better engineers than me can make decisions on the spot that hold up over time. Me, I fuss and fiddle endlessly.

--Ethan

This is precisely it.

-s0nguy

[jyc]

my take:

maybe this isn't a good analogy but think of them like digital cameras...

the scene is your sound source (mic) and the lens is your pre and the CMOS is your converter

looking at specs, different digital camera's cmos can all do 10 million pixels, but some of them does it better than other, why? There're more to a digital image than just resolution, the colour in each pixel, how well those pixels are aligned.

Just like with sound, it's not only about the frequencies, it's how well those are produced. As we know, sound are waves, but very complex wave not of a single shape but a combination of all. Sine waves and Triangular waves sound drastically different, but you can produce the same peak on a frequency response graph with both.

See where I'm heading? Maybe I'm totally, utterly wrong..but this is what I think and hence the difference in converter can be dramatic! It certainly isn't 0.000004%.

Jyc.

[Marjan]

Originally Posted by atticus

Here is a link to the sample page on our website. Thanks!!

MYTEK DIGITAL USA

Every single manufacturer should have a page like that, rather than showing us fancy catalogs.

Someone should start a petition...