[nick_charles]

Quote:

The SNR on the Ref-8 is 116dB and 121dB on the Ref-7, so I don't see how the Ref-8 could be more accurate.
 
It might have a flatter frequency response, but since they do not publish this we can never know, also as far as I can tell the SNR is just for the chip not for the overall system so that may not be as helpful

Quote:

True, but there's no way to measure timbre accuracy of the dac, and has no relationship with SNR specifications.

 

FR would do this but that is not supplied , I would be surprised if it were anything other than flat to ~ 20K though

 
Quote:

No offense Ipod but that SNR spec is meaningless in regards to SQ,  do you listen much above 100dB? 
 
It is true that a SNR above 96db is largely moot but it is a measure of accuracy, added noise is not part of the signal,  if we consider high fidelity to be about accuracy rather than something we like the sound of, even if we cannot tell the difference you can say that one is objectively better than the other but see FR also
 
If you do I hope a hearing aide is in your short term budget.  Let me put it this way Audiogd has more parts cost invested in the Ref8 than any other of their DAC's,  and the comparision shootouts show preference to the 8 over the 7.
 
Is it by any chance newer and thus more hyped ?
 
The Ref8 is the only DAC from Audio-gd without an opamp shown in the analog stage (its the only 100% descrete.)   In China the number 8 is the luckiest most regarded number.  If these don't add up to make it the flagship I think you should at least buy one and form your own opinion

 
and Blind Faith were the world's best supergroup

 

Quote:

Please describe you favorite flavor of ice cream to someone who hasn't tasted it.  Launguage is finite,  it has limits and boundaries,  especially with regard to human perception.
 
Indeed, this is I think the whole problem with the audiophile thingy, non verifiable descriptions that are of limited use to anyone who is not doing the description. But the problem I think is more in psychology than semantics. Someone thinks that an accurate sound is a preferable sound, they buy an item, they like the sound and so they describe the sound as accurate, the next person thinks that a warm sound is preferable they buy the same item like the sound and so describe it is warm. We have no objective scale for warmth or naturalness, but we do have measures of accuracy i.e what deviates least from the input signal, noise, distortion and FR are all measures we can more or less compare directly.
 
Then we come to the second and far worse psychological problem, how do we know we can actually reliably tell two things apart, well if the differences are gross like mono vs stereo no problem , if the differences are big like 10% distortion vs 0.005% very likely, once the differences get smaller it gets harder, but even if the differences are big such as 96db vs 102db we start running into the limits of human perception. Two good quality DACs will have vanishingly low noise, distortion, channel imbalances, jitter and FR deviations beyond audibility (excepting Wadias of course) - so what are the differences we are supposedly hearing ?

 

 

[regal]

That and there is no correlation between a fourier transform plot giving THD to human perception,  surely the ear doesn't break a sound wave into a fourier tansform of different frequencies.  Is it true some researchers found that the ear adds 32% 2nd Harmonic distortion at the ear drum,  that and the fact that there is no scientific model of the perception of sound means that RMAA results get way too much attention.   Even frequency plots are silly if you have ever measured the freqency sweep of a headphone where moving the mic 1mm in a dummy head produces a completely different plot above 9khz. Proving that individual ear geometry has a much bigger impact than a DAC's frequency plot.
 
Our measurements are in the stone age with no correlation to perception.

 

[nick_charles]

Quote:

That and there is no correlation between a fourier transform plot giving THD to human perception,  surely the ear doesn't break a sound wave into a fourier tansform of different frequencies.  Is it true some researchers found that the ear adds 32% 2nd Harmonic distortion at the ear drum,  that and the fact that there is no scientific model of the perception of sound means that RMAA results get way too much attention.   Even frequency plots are silly if you have ever measured the freqency sweep of a headphone where moving the mic 1mm in a dummy head produces a completely different plot above 9khz. Proving that individual ear geometry has a much bigger impact than a DAC's frequency plot.
 
Our measurements are in the stone age with no correlation to perception.

Interesting comments, my background is in Psychology so these things fascinate me, I'll scratch around and see if I can find a reference for that ear thing
 

 

[Nada]

Please can I ask for some advice? Im stuck choosing between the Reference 7 versus 8.
I like a smooth and rich sound signature that can be associated with tubes but I also prefer the dynamics of solid state.
I currently have a Reference 5 with the DSP version from six months back. It has a softer sound signature I like but it could do with a bit more sparkle as long as theres no glare.
Any suggestions will be appreciated. Thanks.

 

[computerparts]

Quote:

Please can I ask for some advice? Im stuck choosing between the Reference 7 versus 8.
I like a smooth and rich sound signature that can be associated with tubes but I also prefer the dynamics of solid state.
I currently have a Reference 5 with the DSP version from six months back. It has a softer sound signature I like but it could do with a bit more sparkle as long as theres no glare.
Any suggestions will be appreciated. Thanks.

You might want to start thinking about options other than audio-gd. Or perhaps you should try out some pure silver (not silver plated) cabling to keep that smoothness but give a little more sparkle with what you already have. What I heard out of the Ref 7 doesn't seem to be what you're looking for. Smooth, yes, sparkle, no, and definitely not rich sounding IMO.

 

[haloxt]

Quote:

Indeed, this is I think the whole problem with the audiophile thingy, non verifiable descriptions that are of limited use to anyone who is not doing the description. But the problem I think is more in psychology than semantics. Someone thinks that an accurate sound is a preferable sound, they buy an item, they like the sound and so they describe the sound as accurate, the next person thinks that a warm sound is preferable they buy the same item like the sound and so describe it is warm. We have no objective scale for warmth or naturalness, but we do have measures of accuracy i.e what deviates least from the input signal, noise, distortion and FR are all measures we can more or less compare directly.
 

There are ways to try to minimize perception fallacies. For example, if someone were to ask the question, Is my setup really neutral, or is it just this recording? He can go about testing that hypothesis by trying different records, and trying those different records on different gear, and deducing an approximate answer to his question. This kind of experimentation is a lot less time consuming than those attempting to statistically prove a listener hears the differences he hears (which in some cases may be impossible for some subjective impressions), but in my opinion, it is a good compromise between blindly throwing out listening impressions and only offering listening opinions that have been scientifically proven.
 
Even what I'm suggesting I sometimes find too time consuming, and I know most people probably don't do it and I take most people's opinions with that in mind. For a long time I would try to disprove any subjective impression I wanted to give, but there are just too many variables I think I have to test and I'm not that keen on critical listening or buying gear to test hypotheses anymore. So now my listening impressions have devolved into something quite similar to what you describe

.

 

[fengshenwee]

 
Quote:

Please can I ask for some advice? Im stuck choosing between the Reference 7 versus 8.
I like a smooth and rich sound signature that can be associated with tubes but I also prefer the dynamics of solid state.
I currently have a Reference 5 with the DSP version from six months back. It has a softer sound signature I like but it could do with a bit more sparkle as long as theres no glare.
Any suggestions will be appreciated. Thanks.

I have the same question a few months ago. There is no right or wrong answer since I dun think u can get the 2 set together and do AB comparison. No one is crazy enough to buy Reference 7 and 8 together. The closest u get is Ref 1 and Reference 8 together. According to him, there is little difference with reference 8 being a little bit, just a bit warmer and forward. Ref 1 being a bit more dynamic. I also read Reference 8 got more bass.
 
I end up with Reference 8 because my current set up is so neutral, I thought some warm will make it more musical and sweet. I am very sensitive to harsh highs so I thought the "warmness" would be nice. I listen to pop and jrock and the recording is not as good as one would like. The audio-gd musical DAC according to Audio-gd seems to be more forgiving than the neutral ones. Furthermore, since I dun use acss, the XLR balanced connection only Reference 8 seems to make more sense. Furthermore, the 100 plus savings, I use it on the Audio-GD DI so it seems to be a better deal. No one that heard my system so far feel that it is warm. It is only warm when u compare to something like Ref 7.

 

[K3cT]

Anyone want to take a guess what are those two op-amps doing in the analog stage of this DAC? DC servo? I thought the analog stage is discrete?
 

 

[FauDrei]

Anyone want to take a guess what are those two op-amps doing in the analog stage of this DAC? DC servo? I thought the analog stage is discrete?

 
Aha - DC servo. All opamps you can find in A-GD mid-high gear use them only for that.

 

[IPodPJ]

Quote:

 
Aha - DC servo. All opamps you can find in A-GD mid-high gear use them only for that.

 
Not in the Ref-8.

 

[vrln]

To my best understanding it doesn´t really matter that much since it´s in the DC servo, but I´d love to hear a REF8 someday too. The only comments I´ve found comparing the two seem to indicate they are quite similar. Oh well, at least I´ll be hearing the diamond differential output in the NFB-11 soon

 

[oqvist]

Back to the Ref9. Listened to some Chopin comparing my GHP and Head One on my LCD-2 again. Piano on the Head One was a bit to soft on the DAC19/Head one combo. Failed to really portrait piano fully.  It was simply not as transparent as the DAC19/GHP combo. However on the Ref 9 the Head One has no such issues. It sound very natural and is as razersharp as you would expect. Tried some movies, gaming and well and the Head One I do feel have a bit better synergy with the Ref 9.
 
 

 

[FauDrei]

Not in the Ref-8.

 
Read again my post: "...all opamps you can find in A-GD mid-high gear..."
 
You can not find opamps in RE-8.

 

[Orangecrush]

Just wondering, what would be the benefit of having two sets of RCA outputs on a DAC?

 

[FauDrei]

To connect for example - main speaker amp and, let's say headphone amp in the same time?