While there are likely measureable differences between the M51 and the Gungnir I would be willing to bet in a blind testing situation over enough trials to account for chance, it would be very difficult to reliably pick the two DACs apart. We all think our “audio-memory” is that good, but that isn’t really how the brain works. We can store certain gross concepts of a sound in memory, but we are unable to actually “warehouse” any sound which ultimately means that once you stop listening to a piece of equipment, whatever detailed sound characteristics you noticed before will have to be re-detected by your brain each time. So when we sell our gear and have a period of time until the new gear arrives suddenly we hear the new gear as being so much better, but how accurate do you really think your auditory memory of your past equipment really was? Is it possible that you don’t actually have an accurate mental representation of that equipment left in memory?
I know when I sold my Audiolab 8200 to fund the Gungnir purchase I had a period of time with no DAC. Sure I believed the Gungnir to sound better, but how confident should I be in my musical memory? There is a huge difference between remembering general sonic characteristics as opposed to the granular level of detail retrieval and musicality we evaluate with gear like a DAC. Does our brain really store a super detailed sonic memory of sound? Why would it do that? Sure we remember voices, but the differences between voices are often striking, and sometimes we hear somebody’s voice and say man did that guy ever sound like Bill. The brain stores general chunks of information, some of the chunks are more detailed than others, but the final, and main auditory picture is put together by the brain while you are listening, it isn’t based on pure retrieval from memory. This implies that we do not have flawlessly detailed sonic maps in our memory from which to draw. I also will acknowledge the factor of individual difference, so there may be some people with more specialized memory circuits who indeed might remember more detail than the average, but these people should statistically be in the significant minority, and it would require properly designed blind testing procedures to confirm.
I know we do sometimes hear detail that we didn’t hear before, and while I think that may be true in some instances, it can just as easily be explained by our attentiveness to our new equipment. You’re primed to detect differences as you have created an expectation in your mind of a difference or you wouldn’t have bothered buying the new gear would you? It doesn’t matter if you have been in this hobby for 40 years and have tons of experience, and feel you don’t have expectations anymore, we all do, our brains are essentially the same and we can’t just turn off evolutionary factors. I believe auditory differences between a budget DAC with a less than stellar analogue section as compared to a very good DAC would be reasonable to say are detectable; however, the differences between two quality DACs would be quite another story unless they were engineered in such a way as to have very different sonic characteristics, which I have to admit is possible, but I don’t know if that is the case frequently. I would have to think that the sonic characteristics people respond to are essentially fairly similar so I can’t imagine engineers taking huge risks and making their gear really, really different sounding.
My point in saying all this is that if you believe you will hear a big difference for those extra hundreds of dollars, then you probably will. But, in my mind the well-designed DACs will probably sound quite a bit the same over much of the sound spectrum. I would love for head-fi to finally set-up some well-designed blind listening tests at a meet. While I am a skeptic of the uber “golden-ear” that can detect minute differences, I am not so full of myself as to think I know the answers, and I could end up looking like an opinionated fool. It actually would be very easy to design a testing protocol and environment and I wonder why this isn’t done at big meets and industry shows?