What's the best deal in a 2 ch. AD/DA unit these days?

[honkyjonk]

Yup,

i like how Apogee converters sound, but I certainly don't have a problem with something quote un-quote more transparent.

Anyway, what do you guys think is the best deal these days, if one were to require 2 ch. of AD and DA. ?

[Ryan Silva]

More transparant than Apogee? Thats a tall order, I wish you the best of luck. Maybe 'RME'.... I haven't listen to them but I hear they do very well.

Good Luck

[honkyjonk]

Naw,

What I meant was, Apogee is fine because I like how they sound, but more transparent is also fine. I'm just going on what various messageboard members say, but Mytek and Lavry in particular are supposed to be more transparent.

But I suppose everyone's MMV.

But what's the best deal?

New? Used?

[I'm Painting Again]

I really like Lavry's sound and features..they are crazy expensive though..

[rjd2]

i called mercenary about ad/da's this week. the guy told me that the lynx aurora held its own against the apogee rosetta, in some instances outshining it, and i had only asked about the rosetta. he was basically trying to sell me on the lynx, which sells for 750 less, so i have to assume someone out there thinks its a decent product. i asked him what i am going to hear from the lynx to the lavry, he said "alot more separation in the tracks".

[honkyjonk]

Huh?

A lot more separation in the tracks? What does that mean? there's never crosstalk between different channels on the converter is there? Funny that people love tape because it glues things together.

Maybe he means the illusion of a wider stereo field.

[I'm Painting Again]

no cross talk and no not wider stereo..I know what they mean but I'm having trouble finding the words to describe it..I think the term some people use, "air space" is what they mean..

[puffpastry]

Maybe 'RME'.... I haven't listen to them but I hear they do very well.

I don't think so. They're really more of a 'mid-grade' converter. They sport the same crappy 4580-based line stage as a Behringer console, and their clock jitter is higher than the Apogee.

the guy told me that the lynx aurora held its own against the apogee rosetta, in some instances outshining it

Wow, I really find it hard to believe that about the Lynx. The Lynx stuff is about the same as the RME stuff. Good converters, so-so clock, so-so linestage.

there's never crosstalk between different channels on the converter is there?

Yes, there is, actually--through the power supply.

The transparency you're talking about comes from the linestages. Benchmark, for example, uses 5532 opamps biased heavily into class A to get its signature sound. In converters like Prism, Lavry, Radar, etc. you're paying for extremely wide-bandwidth, ultra-fast linestages. That's a big part of the 'air' and 'transparency' associated with the expensive stuff.

After that, the two big things affecting converter sound quality are:
1)the clock:
a low quality clock will do lots of damage. One problem is phase cancellations during D/A conversion, giving you a limp, flat sound, and an apparent loss in volume. Digital audio is very famous for being limp and cold. Poor clocking is the reason why. Another problem that arises is the creation of high-order distortions during the A/D.

2)the converter's dynamic range:
low dynamic range gives the music a bit of an edgy, brittle quality, especially when tracking at a higher sampling rate. A high dynamic range gives a much greater sense of 'depth' to the music.

[I'm Painting Again]

there's never crosstalk between different channels on the converter is there?

Yes, there is, actually--through the power supply.

Woah!..I had no idea..learn something new everyday..

how does that happen? (in dumbass terms please)

[puffpastry]

how does that happen? (in dumbass terms please)

I'll do my best to keep it simple, but I'm assuming you know ohm's law.

It typically happens like this:
The digital circuit pulls current from a power supply that is also shared with an analog circuit. The current draw causes the supply to move with each bit. This bleeds into the signal path through an opamp's power supply pin.

We can eliminate (or at least reduce) the problem by using different power supplies for the analog and digital circuitry, or by putting a low pass filter between them. In my personal experience, the use of a circuit known as a pi filter seems to do the trick very well.

There are actually two other ways that crosstalk can occur in a converter circuit. One way is through something known as capacitive coupling. This happens when analog and digital signal lines are run too closely together on a circuit board.

The other way that it occurs is through the ground, or at least through a poorly implemented ground. What happens in that case is that the ground appears as a small impedance to the power as it tries to move from high to zero potential. This makes the circuit ground appear to shift around as current draw varies, and this contributes to crosstalk.

[I'm Painting Again]

hey thanks man..that make perfect sense..

What kind of amounts of crosstalk are people measuring?

seems like it would be a very small amount, on something well designed at least, since you can eliminate it with a filter or making a proper ground that doesn't cause that small load effect on the power..I can see it being tricky I guess due to space limitations to get the different lines far enough apart..

thinking about it..I'm not a tech or anything..seems like space and isolation of components and paths is like so important..I'm surprised I dont see more barrier sorts of thingies inside gear..like for blocking feilds from transformers and stuff like the capacitive coupling your talkin' about..

[puffpastry]

What kind of amounts of crosstalk are people measuring?

It's fairly complex, because we're talking about two entirely different types of signals, and the subsequent effects that they have when bleeding across channels (or within a channel). Really, it's beyond the scope of our simple (and admittedly hi-jacked) thread here, but it's often manifested as a harmonic distortion or signal dependant error. Probably one oversimplified way of looking at it is to say that it creates unwanted noise, at least when we're talking about digital signal cross-talking into the analog signal. So in that respect, I suppose we'd measure it in good old fashioned decibels, huh?

I'm surprised I dont see more barrier sorts of thingies inside gear

It gets expensive to do that sort of thing, and you typically find it only in very high-end stuff. Obviously when trying to meet a lower price point, the designer will have to cut corners. The key, in my opinion, is knowing which corners can be cut and which ones can't. We might not be able to afford 'fences' within our $500 two channel converter, but we could probably afford to use opamps with a very high power supply rejection, or we could probably afford to add low-pass filters in the power supply.

[I'm Painting Again]

Thanks again puff for your insight..it was really enlightening..

BTW I think your black lion stuff rules..and is such a great and helpful thing to our community..

[runrunrun]

puffpastry,

do you have a recommendation for a 2 channel AD DA that isnt ultra expensive? is there one that you can mod?

[puffpastry]

BTW I think your black lion stuff rules..and is such a great and helpful thing to our community..

Thank you. That's very kind of you to say.

do you have a recommendation for a 2 channel AD DA that isnt ultra expensive? is there one that you can mod?

I've been searching for something like this for quite some time. So far, I've not come across anything that feel like would really compete with the mini-me once it's been modified. It seems like all of the inexpensive two channel stuff is pretty bare-bones, and has poor dynamic response. Maybe we should just design one ourselves...