Ok this will be a good thread.

While my "source" cost more than my speakers in my own system, the reasons for placing speakers highest on the chain are born out not only by my experiences. It is also supported by accomplished designers of audio componenets, some of whom I have the luck of knowing, who have had the greatest struggles designing really accurate speakers. Of course you can't have crap in the chain and get there. But if you have speakers that are "voiced" with a particular frequency drop out, you can't get there with the source, either.

Of course, one would need to know what music really sounds like. You are more likely to know that by going to your local symphony rather than your local hifi shop.

Charlie

Danvetc is correct. I use what works with high standards, regardless of price. I found a relatively inexpensive CDP that does everything I expect from listening to live performances. I bought my speakers for little more, but are sublime. Then I had to shell out the big bucks for an amp that can power 1 ohm. I suppose I was in a rare field where I knew exactly what speaker I would use. The amp followed out of necessity. The CDP is just plain great. It is a hold over from a less expensive system.

TWL, if I can hear differences among all speakers, but can't tell differences among CDP's, is it because the speakers are more important for a good system or is it because the speakers aren't good enough to resolve the differences among the CDPs? The answer to that question(no matter which) proves that it is the speaker that is the most important element. If it isn't good enough, then one can't appreciate how good the source is - good stuff in, garbage out.

OK folks,
If you hook-up cheap headphones some Coby stuff or so with descent headphone amp let's say Grado and start swaping CD-players or analogue setups the difference will be much more audiable than if you would change speakers for a pocket CD-player.

There is a point where for a particular room there can't be better speaker and amp and anything invested onto these components will just produce no positive result.

And here we have a debate or arguments what goes first egg or chicken. Everything is variable and needs multi-variable understanding.

I would draw the following curve(s) and describe them by words:
1st point. Let's say I have $100 pocket CD-player,Nad 50W/ch receiver and starting with Boston Acoustic speakers.
2nd point. I upgrade speakers to B&W CDM1 and somewhat shure that it would be better upgrade rather than investing to a new CD-player or amp(meaning and always meaning smaller investment for a better sound)
3rd point. That's where the curve might split or at least take a different direction where non-speaker investment will be more appropriate than spending on speaker?...
4th point. ...might bring you either back to the 2nd or realy towards non-speaker upgrade i.e to the 3rd point.

The so called importance curve of the system components can be represented very similar to the output tube or transistor characteristic as a family of curves. The orts are Performance(vertical) as function of Money Spent(horizontal) with only ONE constant component which is ROOM.
The only exception i guess will be the source especially if it's analogue. While Speaker, Preamp, Amplifier will have the curves(parabolic forms) exactly as shown here similar to output characteristics of bipolar junction transistor are introduced:Page 4 Fig 3 the SOURCE will have a straight line (Performance = C*Money Spent where C is a room constant or let's say tangent of horizontal corner of the function) towards an infinity or the most expencive source component ever produced.

Please, note that there might be exeptions or different even improvised jumps of such curve family(just like in reality) and you might visualize just by only drowing the curve of your previous upgrades that while power amplifier can be considered as if it were an ideal match to the speakers. Thus there can be upto many speakers of a different price range that perfectly matched to the amplifier and will produce significantly larger improvement if the money spent for the amplifier.
On some point source reaches the speaker curve and that's where money spend onto the source will produce much higher results than on speakers or any other components.

Inpep, my answer to that is, that I can hear the differences between the CDPs, even on mediocre speakers. If I can't hear a difference, then I doubt there is much if any difference between them, and one would not represent an upgrade over the other. I realize that it is harder to tell the differences between similarly priced CDPs, because there really is not much difference between them, and the digital source material is so limited, that the primary differences will be fairly expensive to get, such as a quality analog output from the DAC and low-jitter mechanisms. This is why the Linn CD12 and others are so much more expensive.

However the main problem with digital sources is, that even if they get the sound off the disc and into the system, it is far lower quality than a similarly priced analog source. A large percentage of the musical information never got onto the disc in the first place, due to the sampling technology that is used in all digital recordings. A "sample" means that the whole wave is not used, only little sections chopped out on a regular timing structure. If you think that this can give equal performance to replaying the entire wave, then I can't help you with that.

The main challenge of analog systems is recovering the massive amount of information off of the record, and this is why improving your TT can improve the sound so much. No TT has ever been able to totally extract all the info from the groove, but at least the TT gets a continuous waveform to work with. The better it does at recovery, the better the sound is. The real source is the recording and a continuous wave recording with all its warts, is still a better source than a sampled wave recording.

With digital systems the challenge is to make something decent sounding out of a sampled and chopped up waveform that left a good portion of its information on the recording studio floor. Even if it recovers 100% of what it can get from the disc, it is not enough. A five minute comparison of a $6k TT vs a $6k CDP will tell you more about this fact than I could explain in a term paper.

The speakers are an important part of the system, just as all parts are important. I don't take this lightly. But I have done comparisons, and understand the factors involved, and even a modest speaker that you might call "unresolving" will still tell the tale. It might not give the 20Hz lows or even the 20kHz highs, but the musical presentation will be audible, and will reveal the better source.

I have never heard any speaker in any decent audio store that had any speaker that was so bad that it wouldn't show the quality of what was in front of it. And if there is such a speaker, why would anybody who owned it care at all about anything else in their system?

I get the feeling that you are being argumentative for argument's sake. Yes, I agree if you unplug the speaker, no sound will come out. If you have a speaker that makes any decent kind of sound, the front end will be easily discernable.