Opinions on the role of the preamp in a modern single source system

[JackD201]

And of course the other question that must be asked along with this one is what is "good?" It should be very simple and demonstrable. I'm betting we won't be able to keep it that way.

Tim

The demonstrable part is easy at least with the Emmlabs DCC2. Just do what I did. Bypass out into Line in vs Line out into the preamp's bypass. That way you are only comparing the variable analog output stages of both devices. Total snap to AB. Then do a third comparison direct into the amplifier(not quite as easy). Bypass out into Line In won out. It must be noted however that I run very long interconnects from source/pre to amps. Twelve meters per channel to be exact. Many preamps cough up fur balls at these lengths.

With the DP-78 it isn't so cut and dried. You're either at unity gain out or attenuated.

In any case, I'm not a single source guy so my interest was just academic. I mean, I own the darned things so I just had to see what they could do so I tried. In my situation, having a preamp won.

So to answer the question, is it essential, I'd say no, just not as good.

 

[Phelonious Ponk]

The demonstrable part is easy at least with the Emmlabs DCC2. Just do what I did. Bypass out into Line in vs Line out into the preamp's bypass. That way you are only comparing the variable analog output stages of both devices. Total snap to AB. Then do a third comparison direct into the amplifier(not quite as easy). Bypass out into Line In won out. It must be noted however that I run very long interconnects from source/pre to amps. Twelve meters per channel to be exact. Many preamps cough up fur balls at these lengths.

With the DP-78 it isn't so cut and dried. You're either at unity gain out or attenuated.

In any case, I'm not a single source guy so my interest was just academic. I mean, I own the darned things so I just had to see what they could do so I tried. In my situation, having a preamp won.

So to answer the question, is it essential, I'd say no, just not as good.

That tells you what you like, and it doesn't get much more important than that. But you know that's not what I'm looking for. I want to know what effect that preamp is having -- what do you like? And here's the thing: Let's assume that output of the source and the output of the preamp are the same impedance. Let's assume that the volume controls are of equal quality.

What can the difference be?

Tim

 

[JackD201]

Tim,

You know the answer. A line stage is an amplifier, albeit one scaled down in voltage output. I dare say impedance matching (especially with modern solid state as opposed to tubes) is not as important as preamp output voltage and amplifier input sensitivity matching especially when we're talking about long runs like mine. Make that twice as important when bi-amping which will split the output voltage. It is possible to under power your amp's input stage. The effect? Loss of dynamic range, uneven distortion spectra and non-linear frequency response.

We know empirically that robust pre-amplification is less dependent on cable quality as well as length. Back in my pro days we had to resort to distribution amplifiers for the long runs from the FOH board to the active XOs feeding the amps. That was if the active XO's were in the amplifier racks. If the active XO's were in the outboard processor racks then the distribution amplifiers followed them in the signal chain. It's the same thing at home.

I'm not a believer in religiously keeping all cables short. Unbalanced cable can be run 20ft easy without much penalty but the penalty is there. A non-robust preamp output stage will make this more apparent, that being a drop off in high frequencies as resistance increases with length, something easily measurable.

So what do I like? I like the sound of a system that get's out of the friggin' way. Low distortion, linear FR, low coloration, wide dynamic range, wide bandwidth. Under these criteria, having a preamp with the DCC2 and the DP-78 fared better. Without the preamp, the system was lifeless and displayed compression.

Let's disabuse ourselves of the notion that all people that go for preamps do so for color. Many do, many don't. While all components are colored to some degree, there are preamps out there that are designed to be as less colored as their manufacturers can accomplish, mine being one of these. High speed mosfets, mono-mono stepped attenuators, fully regulated power supplies, with a focus on uniform harmonic distortion throughout FR and load as opposed to scrubbed off THD via feedback. I mention this last bit because you will find many THD specs published but rarely will you see THD+Noise vs Frequency and THD+Noise vs Load.

Going back to my first post, if the onboard pre-amplification module can perform as well under these parameters then I don't see the need for another piece in the chain. It's just that with these two units, they don't. I wish they could. It adds a heck of a selling point when the day comes to unload them. That however would be dishonest and that's not the way I roll.

 

[mep]

Do any audiophiles besides digit lovers who are wedded to a single digital source not use preamps in their systems? A preamp is the heart of any stereo system that employs mulitple sources.

 

[JackD201]

Sure. Phonostages with trim exist like the mentioned Steelhead or the EMT JP-66 but they are few and far between. Then there are integrated amps. Some have actual preamp circuitry while most just have attenuators and switchers. Often times you won't see it in the literature. You'll have to open them up to see for yourself.

 

[RBFC]

How about the quality of the attenuator? Wouldn't a higher-quality volume control affect the sound in a positive way? My Krell processor has a "preamp" mode that only uses the volume control, bypassing all other functions just for purity with its excellent attenuator.

Lee

 

[Keith_W]

Let's assume that output of the source and the output of the preamp are the same impedance. Let's assume that the volume controls are of equal quality.

What can the difference be?

Voltage gain. Many sources do not output sufficient voltage to drive power amps - that is, unless you like listening at very low levels all the time. Sometimes you need to drive the power amps above unity gain.

 

[JackD201]

How about the quality of the attenuator? Wouldn't a higher-quality volume control affect the sound in a positive way? My Krell processor has a "preamp" mode that only uses the volume control, bypassing all other functions just for purity with its excellent attenuator.

Lee

Yup.

 

[JackD201]

Voltage gain. Many sources do not output sufficient voltage to drive power amps - that is, unless you like listening at very low levels all the time. Sometimes you need to drive the power amps above unity gain.

Hot damn Doc, you said in two sentence what took me 6 paragraphs.

 

[Phelonious Ponk]

Voltage gain. Many sources do not output sufficient voltage to drive power amps - that is, unless you like listening at very low levels all the time. Sometimes you need to drive the power amps above unity gain.

Gain. Yes. But we wouldn't try to drive an amp directly from a source with insufficient gain. Or a poorly-engineered volume control. So when we're talking, at least when we're talking seriously, about eliminating the preamp, what we're actually talking about is a minimalist preamp as the output stage of the source, taking the place of the preamp.

How about the quality of the attenuator? Wouldn't a higher-quality volume control affect the sound in a positive way?

Actually, no. A higher-quality attenuator would simply affect the sound less.

Tim

 

[JackD201]

Gain. Yes. But we wouldn't try to drive an amp directly from a source with insufficient gain. Or a poorly-engineered volume control. So when we're talking, at least when we're talking seriously, about eliminating the preamp, what we're actually talking about is a minimalist preamp as the output stage of the source, taking the place of the preamp.



Actually, no. A higher-quality attenuator would simply affect the sound less.

Tim

On the first part, EXACTLY! It's about both design and execution. It can range anywhere from an analog or digital trim on unity gain output to something with a robust gain stage with precision attenuators built in. Due diligence is recommended. I can think of three manufacturers that have taken a fully integrated approach (source to amps) that pretty much does away with a traditional preamp: Meridian, Wadia and Behold. To a lesser degree there's Ayon Audio (players work best with some ranges of their amps while some amp ranges may need the added oomph of their stand alone preamps. I'm sure there are more out there and that they too have got gain matching covered. Mixing up their players with other's amplifiers however will probably require more owner involvement.

On the second part, Semantics.

As the saying goes, less is more. Higher quality, less degradation, better fidelity.

 

[AudioExplorations]

Guys, fantastic discussion, this is exactly what I was hoping for.

As we now have some good data from the poll I will join in. Results up to now (taking only into account people who actually tested with and without preamp) are 7/12 (60%) in favor of the preamp for sonic reasons.

I think everything has already been said - I believe the pro-preamp people are either correcting impedance issues or enjoying coloration (to some painful to admit, but why not if you enjoy this!). In a discussion offline with Tim we basically concluded "Who on earth, having spent big money on a preamp, will admit that besides input switching/volume functionality it plays no role sonically?"

Sorry if this is provocative... I am simply after the truth of the matter.

I personally have not done a whole lot of testing but have tested a number of preamps and found that, especially when I switched over to the Medea DAC (with hugely robust output stage) that the trade-off was never worth it.

I do know of someone who has done EXTENSIVE testing (Elbroth over at AA), he posted about the various preamps he had tried out in his system, front end being dCs Puccini/Scarlatti, feeding into DartZeel NHB-108 and Wilson Sasha.

BAT 51SE (siting on the old rack)

ARC Ref 3 (I sold this one after I tried the dCS Puccini direct to my ARC Ref 110 power amp)

Placette RVC passive

Classe Omega

Spectral DMC-30SS (tried it with and without the matching Spectral power amp)

Kondo M-1000 mk2 ($140.000 in retail !)

Jadis JP-80 Gold

Nagra PL-P

Dartzeel NHB-18NS (sorry no photo)

What he wrote was the following:

In general, adding the preamp always resulted in reduced resolution, clarity, imediacy and added grain. They also colored the sound, in one way or the other.

A preamp, beeing an extra component in the signal path, can only substract information. It can also add, but not information, but colorations and phase shift instead. Those may be pleasing to some ears, why not, but then, if you have to add coloration to make your sound more listenable, you have some other serious problems with your system.

Having said that, making the system sound good without a preamp is difficoult. But the rewards are well worth the trouble, IMO.

You need to have a source component with a very high quality attenuation. If the attenuation is in digital domain, you have to take care of source output voltage and amp input sensivity/amp gain, or otherwise, you may end up with using TOO MUCH digital attenuation, which is never a good thing.

 

[Phelonious Ponk]

Mixing up their players with other's amplifiers however will probably require more owner involvement.

Perhaps. I'm not sure that compatibility is as difficult as Audiophiles imagine. A discussion that belongs in one of our many synergy threads.

On the second part, Semantics.

Perhaps again. Very important semantics at least. All tied up in the semantics of essential and in the complexity of designs and the perceived importance of the component in the system. The minimalist view is, I think, more realistic and more productive than the romantic one.

As the saying goes, less is more. Higher quality, less degradation, better fidelity.

A good argument for a simple gain stage and a quality potentiometer.

Tim

 

[MylesBAstor]

Guys, fantastic discussion, this is exactly what I was hoping for.

As we now have some good data from the poll I will join in. Results up to now (taking only into account people who actually tested with and without preamp) are 7/12 (60%) in favor of the preamp for sonic reasons.

I think everything has already been said - I believe the pro-preamp people are either correcting impedance issues or enjoying coloration (to some painful to admit, but why not if you enjoy this!). In a discussion offline with Tim we basically concluded "Who on earth, having spent big money on a preamp, will admit that besides input switching/volume functionality it plays no role sonically?"

Sorry if this is provocative... I am simply after the truth of the matter.

I personally have not done a whole lot of testing but have tested a number of preamps and found that, especially when I switched over to the Medea DAC (with hugely robust output stage) that the trade-off was never worth it.

I do know of someone who has done EXTENSIVE testing (Elbroth over at AA), he posted about the various preamps he had tried out in his system, front end being dCs Puccini/Scarlatti, feeding into DartZeel NHB-108 and Wilson Sasha.

BAT 51SE (siting on the old rack)

ARC Ref 3 (I sold this one after I tried the dCS Puccini direct to my ARC Ref 110 power amp)

Placette RVC passive

Classe Omega

Spectral DMC-30SS (tried it with and without the matching Spectral power amp)

Kondo M-1000 mk2 ($140.000 in retail !)

Jadis JP-80 Gold

Nagra PL-P

Dartzeel NHB-18NS (sorry no photo)

What he wrote was the following:

Actually it was only 50 pct, not 60 pct. in a very limited poll And lots of discussion by people who have never compared the two instances, say even in a unit that offers both passive and active sections at the same time. It also seems interesting that no one has talked about the difficulty of correcting for frequency response aberrations induced using just a volume control.

 

[AudioExplorations]

58.33% actually, I rounded up...

meaningless thought with 14 responses, it does make the point however that there are alot of preamp believers in this neck of the woods.

 

[JackD201]

A good argument for a simple gain stage and a quality potentiometer.

Tim

Indeed. My favorite preamps are at most dual stage affairs with very simple circuits. Always dual mono too. I pass on fancy displays and switching as well as some very useful creature comforts such as programmable gain settings per input. I do however look very closely at how robust power supplies are done as well as how well these are isolated from the audio circuits. In my case I usually end up with two box affairs with DC connected via an umbilical.

As far as attenuators go, there are plus and minuses for all of them be they stepped, tvc or any other kind even optical. I find this not too much of an issue at all.

In the end the trade off choices are made around practical considerations or at least they should be. I'd rather have some added complication if it got the job done rather than have stuck rigidly to a platonic ideal but fall flat on my face for the sake of it.

 

[Keith_W]

Actually guys, I think I should make my position more clear. I have seen all manner of systems. Essentially, a preamp is a gain device that provides proper voltage with volume control at the correct impedance to drive a power amp. I have come across some very fancy DAC's which essentially fulfil the role of a preamp - analog volume control, adequate gain, and correct output impedance - such that the addition of another preamp in the circuit actually colours the sound. DAC's like these essentially have the preamp built into the DAC chassis. In my view - a preamp is highly system dependent and it is not possible to make a generalisation about whether it provides a nett gain or a nett sonic loss. At best, it would be fair to say - with most sources, most preamps provide a nett sonic gain.

 

[MylesBAstor]

OK see where you got the number from; 12 respondents.

 

[Phelonious Ponk]

At best, it would be fair to say - with most sources, most preamps provide a nett sonic gain.

Point taken. Quite literally.

Tim

 

[AudioExplorations]

DAC's like these essentially have the preamp built into the DAC chassis.

This statement is not technically correct. DACs with volume control do not have a preamp on board in addition to regular DAC circuitry.

Here is the typical circuit path with a preamp: DAC chip - I/V stage - output buffer stage (optional) - cable - preamp input buffer stage - analog volume control - preamp output buffer stage - cable - amplifier input = 7 "stages".

Now here is DAC direct with digital volume control: DAC chip with onboard volume - I/V stage - output buffer stage (optional) - cable - amplifier input = 4 "stages". OK, so if the DAC in question uses an analog volume control, that would add one stage = 5 "stages", between the I/V stage and the output buffer stage. So without a preamp we have, at least, one less cable and its connections, and two less active gain/buffer stages.