My little contributions here :
Hi-fi means high fidelity, ie. reproducing music in a way that is very close to the real thing which is heard during a recording or live performance.
In some instances, an understanding of how the composer (by reading the program notes) wanted the music piece to be played is also important. This relates to how the emotion and feeling of the music should be communicated to the audience by the players or singers and whether such elements or sensations have actually been felt.
So here, we have the emotion, feeling, musical details and sense of "live" performance as key attributes to make a music piece sound "great" to our ears. That is how i make judgement to whether the music from headphones or speakers sound fabulous and if so, i'll think the gears are good stuff or having right matching.
So everyone here, how frequent do you attend concerts...whether acoustic jazz (which seldom uses electrical amplication - therefore what you hear is what they play, plus the effect from enviromental acoustic), concerto (must listen in esplanade's concert hall - to hear how the various musical instruments sound in aggregration, and the feelings injected by the players and conductor) and vocals (whether in studio rooms or larger performance theatre) ?. The is the REAL reference point upon which the recordings or music from your sound cards or CDP can be compared. If you don't, then it is extremely difficult to explain to you the meaning of "musicality" or "sonic nirvana" in real sense.
On headphone listening, because the generation of music is coming directly into both ears without any reflective sound waves from the surrounding, the effect is already artificial. Hence, in most cases, music is coming from within the head, not in front of you. This requires psychological adjustments or using x-feed, but still it is different from normal concert listening. However, the important element is whether you can still enjoy and feel the music. And the compensating factor for the loss of frontal soundstage is the increase in details that you should be able to listen to (vis-a-vis the speaker system). Here, the amplification is critical to serve this objective. Also, as there is no environmental "filtering" to the music coming from your CDP, what comes out of the cans is what goes right into your ears. So, where is the "dampering" effect ? The headamp again. It serves to provide certain sonic signature that makes your listening more pleasurable, with description like "less harsh", "warmer", "tight bass", "sweet vocal", "nice timbre"....of course, i wouldn't want to commit that all these attributes are solely the result of right amplication. Instead, i would say that it is the result of combination of the other system components, ie. matching and balance of the source, headamp and the cans.
Findings on the matching and balancing acts for various systems can be found in various review magazines. It has a long history. So this topic is best left to be dealt with in separate occasion. However, one important conclusion coming out of the mix-n-match exercises is that the weakest link (ie. bottleneck) should be ultimately re-located to the most costly item in the whole value chain of the sound system. This would optimise the cost/efficiency leading to the best possible musicality that the system can achieved and involves the few steps of upgrading that we commonly see (called debottlenecking in technical terms).
To me, however, i place a lot of emphasis on the Source. It is the starting point of where the music flows. It is the point where the choice of recording has the largest impact on the ultimate sound output. It is the point where a lot of technologies have been researched and developed to break new frontiers such as SACD, DVD-A and the new hybrid. It is the point where most tweakings (whether on power supplies, resonance dampering, vibration control and EMI/RFI shielding) can be economically done to achieve immediate sonic upgrade. With this importance, i will allocate 50% of my budget to the source, with 30% for the cans and balance 20% for the headamp. However, while headamp takes a backseat on fund allocation, its importance should not be neglected.
And to reduce this portion of budget for headamp even further, i would recommend that you try one of the clones. For me, the Clones RA-1 was a pleasant surprise that blows my mind away (i'll give a short reveiw later on). This clone, bearing the signature of firefox, has such a good match with my Grado RS-2 that it becomes my ritual greet for the morning by spending at least 20 minutes listening thru it on the unique sound of Philip Glass. Talking of decay, i could hear the third or even fourth harmonics from the timebre that it generates. And the sweet vocal from Yang Xiao Ling is forever so magnetising. Wow...i just simply can't help but to shelf other headamps away.....
So here i am, with little knowledge on the latest DAC technologies or technical terms or specs ...eg. ?? AKM4381 a 24/192 dac on 110dB dynamic range, ?? Lynxtwo on CS4396, 120dB DR, 24/192,-100 dB THD+N, this is my little contribution on the topic, ablaze. In fact, I can't differentiate between good or bad DACs by sighting. But let me listen to a marvellous piece of music with good recording, and i can certainly tell you the truth. Nothing but the truth, for headfiers, audiophiles and the like........I am on my way to sonic nirvana.