[Frihed91]

You all make it seem so logical, and so it is important to start with objectives in mind and know the constraints that limit you...BUT once you start to implement you plan.... throw in the possibility that having bought components A, B and C in that order so that A matches B and C matches A and B, what happens when you audition component D and it sounds not so good with A and B, but oh so much better if you switch out A and B for A' and B', gear you had never considered before?  Things like this happen as a result of learning by doing recursively rather than simultaneously (which is impossible anyway), and the market doesn't that work that way.  This is what some people wrongly call upgrade fever.  (Ask Phil Lamm to explain the difference bewteen problems that can be solved recursively vs. simultanously)

I think that this is wrongly named.  The majority of people who upgrade are doing it as a result of learning more about their own preferences for sound and how various technologies, brands and models fit together with that.   I think that is also what makes the process/journey so much fun, because we are always finding out new things, often randomly, and we have to figure out how to fit this all together in our heads, our hearts, our checkbooks, our listening rooms.  It isn't easy.  And then get the wives to agree....

I say be preapred for surprises.  Don't fear them, enjoy them, just as you learn to love a new composer or the solution to a set of higher-order differential equations!

[woodsyi]

I say be prepared for surprises.  Don't fear them, enjoy them, just as you learn to love a new composer or the solution to a set of higher-order differential equations!

Absolutely.  How many times, do we hear, "I didn't know I was missing ____ until I heard it."  Rational guideline is for planning but emotional response will dictate the actual course system building.  Or should I say bodily response as in crapping inducing bass, goose-bumping vocals, toe-tapping pace or hair-on-the-back-of-your-neck raising screecher of treble? 

[Bob Reynolds]

Interesting comments from all. I tend to agree with bits and pieces from many.

I personally think one should always listen nearfield -- about 7 feet away from the speakers. Thus, the room has less impact.

I was surprised that no one suggested the obvious solution to picking amps and speakers -- active speakers. Let the designer implement both.

So if I was starting over today, I'd pick active speakers first and those would be ones designed to be mated with a subwoofer.

Next would be an active preamp and then any decent digital source.

I don't believe in exotic cabling (Belden works just fine for me) nor tweaking, so I'd take that money saved and apply it to electronic room correction.

[JLM]

I fully agree with you Bob regarding active speakers and letting the speaker manufacturer make the selection for you, with the caveat that many "speaker guys" are so quick to dismiss all other links in the audio chain.  After trying about 8 different amps I finally listened to Bob Brines, the guy who designed/built my speakers, and went with his amp suggestion.  Would have listened to him sooner, but he's a bit of one of "those" speaker guys.   

A few years ago Paradigm offered active and passive versions of their Studio 20 and 40 (standmount 2-ways) speakers, actives being about $800 more.  There was no comparison.  The bass that came out of the small Active 20s blew people away.  IMO it out performed the more expensive large floorstanding Studio 100s. 

This is one of the reasons why single driver speakers and powered subs can work so well.

[PhilNYC]

(Ask Phil Lamm to explain the difference bewteen problems that can be solved recursively vs. simultanously)

Yes, I was just thinking about this the other day...?   

 

[sts9fan]

I would absolutly start with cables first!!

[Frihed91]

What would you hook them up to?

Just a thought

[MaxCast]

What would you hook them up to?

Just a thought

The AC outlet.

[JohnR]

I'd start with the amp. IME that's the hardest part to get right...

How do you pick an amp if you don't know what the speakers are?

Would you simply get an amp that would be powerful enough to mate with almost all speakers out there?

Logically, this argument goes both ways. How would you pick a speaker if you don't know what the amp is? Would you simply get speakers that are sensitive enough to mate with all amps out there?

In practice, I don't think anyone is going to buy either one first without devoting any thought to the other. It's just that it's been my experience that the amp is harder to get right than the speakers... and should therefore get sorted out first, if there's going to be an order about it (which is the title of the thread).

[Steve]

I don't think I would choose the speaker first, maybe a combo of amp/speaker. I have just seen too many subpar sounding speakers to let the speaker dictate the amp.

Actually, for separates, if I knew a preamp that was a "perfect" straightwire, I would choose the preamp first. Then the amp/speaker/source/etc.

[JohnR]

Actually, for separates, if I knew a preamp that was a "perfect" straightwire, I would choose the preamp first. Then the amp/speaker/source/etc.

That's why I suggested a passive in my earlier post. Since it's harder to get "wrong" than active.

Again, this opinion is of course based on my own experience. Everybody's different

[Steve]

I see your point, the integrated/speaker combo (preamp stage incorporated into the basic amp). At least the active 1st stage is partially synergized with the amp.

But the first stage of the integrated isn't really fully optimized in and of itself, in any integrated that I have seen. Plus there are still the problems:

1) Frequency dependent sonic feedback through the power supply from the output and drive stages to the first stage.

2) And we still have the problem of the passive driving the amp properly. As the frequency increases, there is drive current necessary to charge and discharge the IC capacitance and amp input capacitance, which the passive cannot supply.

The only solution I can see with an integrated is to incorporate the volume control  in the integrated itself. At least the IC capacitance (we don't need one now) won't be a factor.

I guess it is pick one's poison. It's a tough question Earlmarc asked, to actually build an audio system from the ground up.

Maybe audition as many complete systems as possible and go from there?

[kfr01]

It takes so long to get all the pieces together, and even longer to tweak a system for maximum performance. If I were to do it all over again, were would I start? First, I think I would find the speakers that fits my taste. I then would  trial numerous amplifiers to mate with the speakers until I found the one that works best with the speakers. After that, I would seek a transparent preamplifier that allowed the amplifier to show its merits. Next, would be the digital source. The last components would be cables, conditioners, and tweaks. How would you approach starting over?

dedicated room (if possible), speakers, room treatment, source, amp, preamp, all others.

[kfr01]

I'd start with the amp. IME that's the hardest part to get right...

How do you pick an amp if you don't know what the speakers are?

Would you simply get an amp that would be powerful enough to mate with almost all speakers out there?

It's just that it's been my experience that the amp is harder to get right than the speakers... and should therefore get sorted out first, if there's going to be an order about it (which is the title of the thread).

I find it hard to believe that a component with distortion from 20hz to 20khz measured in percentage numbers of less than 1% (even $300 denon receiver does this), could possibly be "harder to get right" than speakers --- which are wildly varying in bandwidth, thd, and just about every subjective and objective variable one can shake a stick at. 

You must have demoed some wild amplifiers in your day---or had perfect speakers to begin with. 

Edit:  But then again, I don't have to believe it; its your experience, after all. -  Cheers! 

[rustydoglim]

We published an article specifically for this topic. At that time we haven't launch the S-9 speaker but our position is still the same.
http://www.nuforce.com/recommend/sys-building-guide1.htm
Well, this is a bias point of view from a manufacturer but I think there are many good points worth considering (feel free to ignore some of the marketing statement in there).
You should also read the detail review of Ref 9 from the non-commercial (ie no ad $ influence) TNT-Audio article which gave you a very good understanding of an amplifier's role in the system.
http://www.tnt-audio.com/ampli/nuforce_ref9_e.html

Jason from Nuforce

[JohnR]

I find it hard to believe that a component with distortion from 20hz to 20khz measured in percentage numbers of less than 1% (even $300 denon receiver does this),

Well, I mean... are you suggesting that one should pair a $300 receiver with very expensive loudspeakers because they have comparable THD numbers - ?

[JohnR]

Not sure if this is taking the thread off track, but here's a distortion plot of the (highly regarded) Seas Excel 17cm driver:

http://ldsg.snippets.org/graphics/seas/e0017dist.jpg

Distortion is 0.2% - 0.3% in its best range, and above 1% below about 100 Hz.

For whatever reason, you can hear "through" all of that to what the amplifier is doing.

[Frihed91]

I think: A preamp plays a bigger role in making music than an amp

[kfr01]

I find it hard to believe that a component with distortion from 20hz to 20khz measured in percentage numbers of less than 1% (even $300 denon receiver does this),

Well, I mean... are you suggesting that one should pair a $300 receiver with very expensive loudspeakers because they have comparable THD numbers - ?

   of course not!   I'm talking about the order of magnitudes of differences between different amplifiers and between different speakers.  I suggest that the gross difference means your suggestion (that amplifiers are hardER to get right) is dubious to me.

[kfr01]

Not sure if this is taking the thread off track, but here's a distortion plot of the (highly regarded) Seas Excel 17cm driver:

http://ldsg.snippets.org/graphics/seas/e0017dist.jpg
Distortion is 0.2% - 0.3% in its best range, and above 1% below about 100 Hz.

For whatever reason, you can hear "through" all of that to what the amplifier is doing.

Alright. 
1) That is a very fine driver.  Many good speakers do not use drivers that expensive.  This actually supports my contention.
2) Take that driver and look at its frequency response graph.  Compare it to any competent amplifier out there.  Here it is:  http://ldsg.snippets.org/graphics/seas/e0017fr+imp.jpg
3) I wish I could find the waterfall plot, showing that even the finest speakers are worse than even mid-fi receivers in the time domain as well. 

My point wasn't that one specific number makes a denon receiver what everyone should buy.  Frankly, I'm offended that you read my post that out of context.  My point was this:  amplifiers have gotten to the point where good engineering will create a very linear product with amazingly small variations between amplifiers in every objective variable.  The differences between speakers is orders of magnitudes greater.  Thus, speakers vary much more, and are likely "harder" to get right.