Using Stan's DAC in my studio: D/A converter comparison

[foxysounds]

I've been using Stan's TC-7520 in my hifi for some time now and have been very pleased with the improvement it's provided. Now I want to try it in my studio and see if it gives a similar improvement to my monitoring.

In order to connect the output of the TC7520 to my mixing desk I'm going to need to buy some cables. The audio outputs of my current audio card are balanced TRS jacks and they are also connected to balanced inputs on my mixing desk via TRS jacks. The desk will accept unbalanced inputs too so I'm going to need to buy a couple of mono RCA phono to unbalanced TS jack leads. If the test is going to show the TC-7520 to its best and give a meaningful comparison with my current audio card then the cables will need to be decent quality ones and therein lies the problem. Any money I spend on cables will be wasted if I decide that Stan's DAC is no better (or not sufficiently better) to warrant the outlay for a second one.

So before I spend any money I wonder if someone can tell me whether I should expect to see an improvement in quality with an unmodified TC-7520 (using the PCM1716 DAC) compared with the D/A conversion on my audio card which uses AKM AK4393 D/A converters.

Many thanks,

Simon.

[StanleyB]

I haven't yet heard a soundcard that sounds better than a good quality stand alone DAC. And certainly not one that sounded better than the TC-7520.

But if you get the TC-7520 modified to the Caiman spec., that would open up a clear lead between the two DAC options you now got.

Stan

[foxysounds]

Ah, but this isn't just any old sound card :-) It's an M-Audio Delta 1010. Instead of the DAC being on a PCI card inside the PC it's in an external 1U rack-mount case with it's own power supply. It costs about 300ukp new with 8 analgue inputs and outputs and digital I/O via S/PDIF but no pre-amps.

In other words it's been well implemented which is why I was wondering how the DAC chips themselves compare (AK4393 vs PCM1716) before I spend money on the cables to try it for myself.

Thanks,

Simon.

[Codifus]

I think I can offer some insight. The AK4393 is a delta sigma DAC. It has been said on some audio forums that delta sigma DACs have a sound that can be described as very polite, almost too polite. They continue to suggest that delta sigma DACs tend to substitute frequency resolution for amplitude resolution. Once I heard that description, I realized that they were describing my DAC very well; an EMU 0404-USB. It has an AK4396 DAC within. As I've grown used to my 0404, I've found that the sound is very very smooth, intoxicatingly so. Compared to my recently purchased TC-7520, the 7520 beats it in terms of better S/N (blacker backgrounds), and higher resolution. The TC-7520 actually separates the instruments. With my EMU, I hear the instruments clearly. With the 7520, I not only hear them clearly, but can pick them out in space. The violins are over here, the female vocalists are a few steps back within the soundstage etc. And this is the regular, non-Caiman TC-7520.

Keep in mind that these comparisons are with my setup; Itunes on a Mac (in a bit perfect setup) feeding an airport express which optically feeds the EMU 0404 or the 7520. 16/44.1 audio data only.

Aslo, because the emu-0404 USB is an economically priced portable mixing console for the audio professional which just so happens to have a great DAC inside, there may have been compromises in its design, and that may have contributed to its less black backgrounds etc, as compared to the TC7520, a device that was designed to be one thing and one thing only: a DAC.


The 0404 has one redeeming quality; it still sounds the smoothest in my setup, but that advantage, because its smooth sound is only a wee bit better, is insignificant compared to the other qualites. Because the 7520 beats the 0404 in all the other aspects, and it does so quite handily, I'm getting ready to retire my 0404.

If you are an "imaging be dammned" type of guy and want the smoothest frequency response, then the delta sigma type DACs may always be a winner. If you want imaging as well as good sound, well, you know my answer

CD

[foxysounds]

David - many thanks. Tons of good info and lots to think about there.

I would suspect that the Delta 1010 has a better implementation of the DAC than your Emu 0404. However, it's also a slightly different chip (does lower number mean lower quality? I expect so).

I think that like you I would probably prefer a more detailed sound than a more smooth sound - particularly for monitoring purposes where accuracy is everything.

Of course, there are lots of external DACs around that are designed specifically for studio use. They tend to be rack mountable and have balanced inputs and outputs which is great. Unfortunately none of them are anywhere near the low cost of a TC-7520 and I also know from Stan's own comments that he designed the 7520 to sound neutral rather than nice which is exactly what you want in a studio.

The only thing that stops me going out and buying the cables to try my 7520 in the studio is the fact that a pair of decent cables for it will cost me 35ukp or so (price from Mark Grant cables) and if the difference is negligable (or if I prefer the DAC in the 1010) then I doubt I will never use the cables again.

I'll sleep on it :-)

Thanks again David.

Simon.

[DaveK]

The only thing that stops me going out and buying the cables to try my 7520 in the studio is the fact that a pair of decent cables for it will cost me 35ukp or so (price from Mark Grant cables) and if the difference is negligable (or if I prefer the DAC in the 1010) then I doubt I will never use the cables again.

Simon.

Hi Simon,
Go on and buy 'em - if you find that you will never use them again you can always sell them on forum or eBay, there's always a market for good phono leads, IMHO.
Cheers,

[Labarum]

EMU 0404-USB. It has an AK4396 DAC within. As I've grown used to my 0404, I've found that the sound is very very smooth, intoxicatingly so. Compared to my recently purchased TC-7520, the 7520 beats it in terms of better S/N (blacker backgrounds), and higher resolution. The TC-7520 actually separates the instruments. With my EMU, I hear the instruments clearly. With the 7520, I not only hear them clearly, but can pick them out in space. The violins are over here, the female vocalists are a few steps back within the soundstage etc. And this is the regular, non-Caiman TC-7520.

Thanks. I have been looking for a comparison with the EMU 0404-USB.

Some folk rave about it on another forum.

How about a comparison with the DAC in a Logitech Tansporter? Same DAC chip? But a lot more money.

[Chippy_boy]

Ah, but this isn't just any old sound card :-) It's an M-Audio Delta 1010. Instead of the DAC being on a PCI card inside the PC it's in an external 1U rack-mount case with it's own power supply. It costs about 300ukp new with 8 analgue inputs and outputs and digital I/O via S/PDIF but no pre-amps.

In other words it's been well implemented which is why I was wondering how the DAC chips themselves compare (AK4393 vs PCM1716) before I spend money on the cables to try it for myself.

I can't answer this specifically, but I can offer some experiences which may be useful.

I purchased an M-Audio Audiophile 192 some while back for a HTPC. My aspiration was to feed analogue out from the 192 instead of using the dac in my Tag AV32R to do the conversion. I had hoped that because the M-Audio card was based on the newer AK4358, and has decent op-amps, it would sound better than my Tag.

I was very sadly disappointed. It was woeful compared to the Tag. I don't have a particularly refined ear for these things and I really struggle to tell the difference between lots of things. Cables for example pretty much all sound the same to me. I mention this to give you some sense of scale for how difference to the Tag the m192 was. It was *hugely* worse. Like walk into the room and think "Yuck" within 5 seconds. The sound is dull, lifeless, closed in and just horrible.

Yet the Beresford Caiman by contast, is *better* than the Tag. Not by miles - and perhaps not by as much as I might have expected. The two are close. But the Caiman definitely has a deeper, tighter bass and a little bit more "openness" to the sound. It's clearly superior.

The Beresford would *completely* blow my M-Audio card out of the water.

I don't know how similar the M-192 and the 1010 are. They do use different dacs and maybe the 1010 would be better? But I can't imagine it could be that much better?

[foxysounds]

Well, I bit the bullet and bought some cables to try this out (I found some slightly cheaper ones that still seemed to be well made with decent quality components if maybe not quite as high quality as the Mark Grant ones I mentioned earlier).

I set up a mix in my DAW with everything routed through a single group channel which I could then route either to a pair of analogue outputs or the stereo S/PDIF output on the Delta 1010. The analogue outputs were then connected to a stereo input on my mixing desk and the digital output was connected to the TC-7520 via a Mark Grant digital cable. The line output of the TC-7520 was connected to a second stereo input on the desk. The main outputs of the desk were connected to my Adam A7 powered near-field monitors.

As the audio outputs of the Delta 1010 are balanced and the outputs of the TC-7520 are not, I set the relative levels of the faders on the desk by ear until I got the volume levels subjectively the same and then proceeded to listen to a number of my mixes while switching between the two different outputs.

I could tell no difference what so ever. Not even a tiny bit. In fact, they sounded so alike that I had to keep muting and unmuting channels on the desk to prove to myself that I really was switching outputs on my audio card. I even plugged my Beyerdynamic DT770 Pro headphones into the headphone output of the desk and really tried to focus in on details like reverb tails but I still couldn't make out any difference at all.

Now, you may be wondering whether I would have been able to tell a difference if I had been listening to commercially recorded material being played through the DAW and it would have been an interesting test if I had thought to do it. However, in the end it would have been academic because that's not what I use my studio for. If I can't hear more detail in my own mixes then what's the point?

The other quesiton that springs to mind is whether or not the digital output of the Delta 1010 is bit-perfect (using ASIO drivers) and to be honest I don't know the answer to that question. However, I do know that the way I tested it was exactly the way I would use it in a mixing situation so again it's academic in the end.

I guess I don't need new converters for my studio any time soon then :-)

Simon.

[StanleyB]

How does your mixing desk compares sound wise against a preamp? Have you got one of them with all the bells and whistles like tone controls etc?

Stan