What should be the first component when upgrading?

Wow,

Lots of advice and opinions. Ok here it goes http://www.slimdevices.com/pi_squeezebox.html at $249.00 either connected to your powerbook via ethernet or wireless is the way to go for your front end. End of discussion the analogue out far surpasses the Roku. It will play apple lossless and your existing MP3 files as well as WAV. Forget the integrated for now and upgrade the speakers. I would go with a pair of used Dynaudio Audience 42 if you can find them.

My desk system at work is as follows.
Denon CD Rreceiver UD-M31
Dynaudio Audience 42.

This is a great little system that puts me in the speakers are the most important camp.

Good Luck in your search.

I agree with the source. Improving the speaker is fine, but keep in mind the speaker can only reproduce the signal it's being sent. Improve the signal, you improve the sound....everytime. If the source, pre-amp, amp is the problem, changing to a better speaker will merely more clearly reveal the faults of the source, pre or amplifier, making the system sound worse. I seen that happen time after time as a retail audio salesman. Source first. PS: In my opinion, the reason the Grado headphones sounded so good is they are virtually void of phase shift, which most speakers have tons of.

How does one judge the sound quality of a source component without speakers/amp that will resolve the info being fed? Without headphones and an excellent headphone amp it's not possible. One can select a source component known to be resolving based on reviews and user opinions, but it's not possible to judge the component's sonics or compare them against other source components.

Catch 22.

Right on! That's why "source first" is wrong in this instance. And "chain only as good as it's weakest link" is faulty thinking, as I've argued before. It ain't a chain, it's a system.

I don't know that I am steadfast in that sense, because overall I think 'balance'in the chain, though if pressed I'd say source. The reason for the source-first camp is the old adage; garbage in = garbage out. If you start with garbage as the source, there is absolutely nothing down the line you are going to be able to do other than make the garbage smell a bit sweeter, but it's still garbage when it comes out the end. So if you take a source that is not capable of nuance, that is fatiguing, lacking in detail and soundstage, and harsh in the highs, I don't care what you put in the rest of the chain...you can have speakers that look like screen doors, or public waste cans, or giant horns from Dr. Seuss...you can have amps with enough power to light a city, preamps with glowing NOS tubes from the 1950's, and cables that cost enough to feed a family of four for a year; you can have all this stuff in place and it can be working in perfect synchronicity in a room that resembles a miniature world class concert hall and is acoustically perfect..you can add your clever lil' clocks and magic bricks and lead-weighted everything...you feed that crappy source through that otherwise perfect system and you will be treated to harsh highs and no soundstage and you will be fatigued from listening to it at length. You feed the signal from a Close-N'-Play turntable throuth that system and it's not likely to bring any smiles other than smiles of mild amusement at the novelty that you made it sound a lot better than it should. But it still is not going to be something you'll want to keep listening to. Reverse the situation and put a world class front-end with a world class middle components, and crap speakers and I'd bet you could listen and be amazed for a whole lot longer than compromising the source. These are extreme examples to illustrate the point, but the fact is it's all connected in a chain and you should go for balance and strength throughout. That said, I do think that speakers do make a big difference in the kind of sound your system produces...the sonic signature perhaps. This is an endless debate, repeated over and over here. There's my .02 lincolns again.

Marco