[muad]

I decided to post this here in sound science as I would prefer a logical approach to fixing my problem. I figured you guys could perhaps point me in the right direction.

 

So I just picked up an asgard 2 today and was planning on getting rid of my O2/Odac combo. The asgard 2's posted specs should make it technically as transparent as the O2 amp. I went with the asgard 2 for functionality as I like having a preamp and big/smooth volume pot. The source for the Asgard 2 was a seperate odac I was using for my speakers. The only worry I had was that the Asgard 2 wouldn't be as transparent as the O2. When listening to it, I volume matched both amps and used my Q7O1s (eq'd the bass to match the K712 I am about to buy). I noticed a difference. The Asgard 2 sounded noticeably more etched, sibilant, less smooth and less coherent. It didn't sound right.

 

So I decided to run some tests with a cheapo mic in the earcup. I placed the mic in such a way that there was minimal movement between tests to maintain some semblance of accuracy. What I got was a measurable difference in the frequency response, which I was able to repeatedly produce. My immediate assumption was that the Asgard 2 was a piece of ****. I tried it a third time with a lower volume and the result was the same.

 

I was originally going to come on here and bitch about the Asgard 2, but I though I would cover all my bases. The only other variable was that both units were using different sources. The Asgard 2 with a seperate ODAC and the O2 amp with an internal ODAC. So I used the external ODAC and repeated the measurements for both the amps... Lo and behold the measurements of both amps looked identical to the previous Asgard 2 measurements. So essentially my external ODAC sounds etched, sibilant, less smooth and less coherent relative to the ODAC inside my O2 amp. Well ****... I wasn't planning on getting a new dac as well. The whole point was to have a good clean source with peace of mind, now I am not sure if the etched sounding one or the smooth sounding one is more true to the source... stupid objectivism.

So now I'm not sure where to go from here. Do you any of you guys know what's happening? Is there anything else I should check?

 

O2 vs Asgard 2 - after warm up 

 

Did a second one as I thought it was an anomaly 

 

Here is one at a lower volume 

 

Now the O2 and Asgard 2 from the same external odac

 

[stv014]

The whole point was to have a good clean source with peace of mind, now I am not sure if the etched sounding one or the smooth sounding one is more true to the source... stupid objectivism.

 
You could measure the frequency response on the output of both ODACs, and possibly also on the amplifiers, and on the ODAC outputs loaded with the amplifier. Maybe you have an ODAC that is defective in some way.

 

[MaciekN]

The scale on the left is very squished, could you write what kind of difference there is in dB?
 
If we assume that ODAC has consistent quality there are two variables that I can think of. Firstly, your ODACs differ in grounding, one has a separate enclosure and connects to Asgards enclosure throu shield or ground wire, a cable in general, so you get two enclosures with different potentials and more than one return path for currents. The second difference would be the analog cable itself but there would have to be a very considerable amount of capacitance to roll off the treble, or some capacitance and a very high input resistance on the amp.
 
You could try and measure with RMAA, even with an analog input on a PC motherboard, the outputs of both ODACs without any enclosures, one after the other so they would be connected and grounded the same way.

 

[JAG87]

  The scale on the left is very squished, could you write what kind of difference there is in dB?
 
If we assume that ODAC has consistent quality there are two variables that I can think of. Firstly, your ODACs differ in grounding, one has a separate enclosure and connects to Asgards enclosure throu shield or ground wire, a cable in general, so you get two enclosures with different potentials and more than one return path for currents. The second difference would be the analog cable itself but there would have to be a very considerable amount of capacitance to roll off the treble, or some capacitance and a very high input resistance on the amp.
 
You could try and measure with RMAA, even with an analog input on a PC motherboard, the outputs of both ODACs without any enclosures, one after the other so they would be connected and grounded the same way.

 
Bingo.
 
Scientifically, the shortest path for line level analog signals is always better. This is why I have never owned a stand alone DAC or a stand alone amp. The two should always be integrated.

 

[mikeaj]

  Scientifically, the shortest path for line level analog signals is always better. This is why I have never owned a stand alone DAC or a stand alone amp. The two should always be integrated.

 
On the other hand, you need a loooot (a technical term) of capacitance to really move the needle there, and moving the digital stuff closer to the amp can induce some interference (noise) there. Generally it shouldn't really matter, but sometimes you run into corner cases.

 

[superjawes]

Scientifically, the shortest path for line level analog signals is always better. This is why I have never owned a stand alone DAC or a stand alone amp. The two should always be integrated.

Scientifically, you do this to minimize resistance and avoid transmission line effects. However, you won't see the latter unless you have a really long wire or a really high frequency signal (which is plenty inaudible). Resistance is easy to workaround.

So while you might technically get better line resistance from an integrated DAC/Amp, you are also more likely to pick up audible noise caused by digitial switching components. Can you avoid this with a combo DAC/Amp? Yes, but it requires some extra design flair to isolate everything.

 

[muad]

Thanks for the advice. 
 
Just to be clear I can run the tests from the odac out to the mic in? I will need help to interpret the graphs after...
 
EDIT: I already eliminated the cables as I used two different RCA cables during the measurement where I used the external odac through both amps. 
 

  The scale on the left is very squished, could you write what kind of difference there is in dB?
 
If we assume that ODAC has consistent quality there are two variables that I can think of. Firstly, your ODACs differ in grounding, one has a separate enclosure and connects to Asgards enclosure throu shield or ground wire, a cable in general, so you get two enclosures with different potentials and more than one return path for currents. The second difference would be the analog cable itself but there would have to be a very considerable amount of capacitance to roll off the treble, or some capacitance and a very high input resistance on the amp.
 
You could try and measure with RMAA, even with an analog input on a PC motherboard, the outputs of both ODACs without any enclosures, one after the other so they would be connected and grounded the same way.

 

[muad]

Hey guys,
 
So I talked to jds labs and they are going to test both units and let me know whats up. It is probably best this way. Thanks for the advice, I will let you know whatsup

 

[OICWUTUDIDTHAR]

thx, pls keep us posted dude, I am interested in this

 

[MaciekN]

 

Yes, if you haven't shipped both units to JDS yet then you can test them directly with mic input, no headphones in between. Line in would be tiny bit better, because mic in will probably get overloaded by a 2V RMS signal, so you would have to significantly reduce the recording volume but thats OK coz we are looking for relative differences, not absolute values. Line in is designed for line level signals, so there would be less fiddling with it, mic in is also fine.
 
You could use the free Rightmark Audio Analyzer, its extremely easy to install and set up.
 
This way in 5 minutes you could check if they measured different, gives us an update, and then ship it to JDS

 

[muad]

Yeah I was going to do that, then I got a message from JDS and I decided on taking the slacker route... but you are right. I will give it a whirl tomorrow  

 

[muad]

So I am having some problems with RMAA, when I set my levels it says "high distortion in the recorded signal", but if I lower the volume it says the level is too low. I am using the line in on my xonar dg...
 
edit: I gave up on rmaa measurements, just doesn't want to play nice with my system. Another thing I found was that when I listened to etta james - at last, there is audible distortion when I plug with my external odac into the O2, whereas that distortion is gone with the internal odac. It wasn't audible with the asgard 2 amp, maybe the asgard isn't as resolving? not sure. Anyways, I think I have to have jds labs look at it and see what they think.