Best Preamp - NO preamp... (?)

Simply having enough voltage is not the whole story.
There are so many variables (including listening preferences) that, as others here have pointed out, there really is no one answer.
In addition to having enough voltage to drive the amplifier, impedance matching is also important with a low output impedance (from the passive device) feeding a high input impedance (to the amplifier) being desirable.  The length of the IC cable between the preamp or passive device and the amplifier(s) also affects this.  
You cannot lump everything that has gain into one "preamp" basket because there are many different designs.  When folks say that preamps are no longer needed, I believe they are mostly referring to the high gain devices that were common 30 years ago.
The optimal set-up for a passive preamp is between a source with sufficient voltage output to drive the amplifier, and an amplifier with sufficiently high input impedance to not be negatively affected by the output impedance of the passive device, and finally for the connecting IC cable to be sufficiently short to not have an effect.
Even when these things seem appropriate, there are many who still prefer sending their signal through an active "preamp."
The issue is further complicated in that all "passive" devices are not the same.  Some passives are simply a volume control, including a pot or perhaps a discrete resistor type control, and others use transformers, autoformers or even LDRs (light dependent resistors).  The resistor based volume controls seem to be most affected by impedance issues.   I own two resistor-based passives, a Goldpoint single input unit and Endler attenuators which connect directly to the amp so taking the length of the IC cable out of the equation.  Both use discrete resistors and both sound very good but need careful impedance matching between the source and the amp.   I have also had the Acoustic Imagery JaySho here for an audition (this is the same design as Chapman's Bent Audio TAP unit, but repackaged) and while it may open up a wider range of equipment that will work well, IMO there are sonic trade-offs so I wouldn't necessarily say everyone would like the TAP unit better.
George posted a portion of a Nelson Pass quote previously in this thread that was taken from Pass' discussion of his DIY "buffered" preamp.  A "buffered" preamp is generally considered to be a no-gain device with an active electronic circuit that reduces the output impedance.  Here is the other half of Pass' quote;

Is impedance matching an issue? Passive volume controls do have to make a trade-off between input impedance and output impedance. If the input impedance is high, making the input to the volume control easy for the source to drive, then the output impedance is also high, possibly creating difficulty with the input impedance of the power amplifier. And vice versa: If your amplifier prefers low source impedance, then your signal source might have to look at low impedance in the volume control.

This suggests the possibility of using a high quality buffer in conjunction with a volume control. A buffer is still an active circuit using tubes or transistors, but it has no voltage gain – it only interposes itself to make a low impedance into a high impedance, or vice versa.

If you put a buffer in front of a volume control, the control’s low impedance looks like high impedance. If you put a buffer after a volume control, it makes the output impedance much lower. You can put buffers before and after a volume control if you want.

The thing here is to try to make a buffer that is very neutral. Given the simple task, it’s pretty easy to construct simple buffers with very low distortion and noise and very wide bandwidth, all without negative feedback.

Link: https://www.passdiy.com/project/preamplifiers/b1-buffer-preamp

Other manufacturers have also found the no-gain to low-gain approach to work well, such as Steve McCormack who originally offered no-gain or +6dB options on his highly regarded VRE-1 preamp, which is basically a very high quality buffered circuit.  The +6dB option is standard on his current version of the VRE-1 with this relatively low-level of gain resulting from high quality output transformers he uses, and not the active circuitry.   Steve has found the power supply to be very important to the quality of sound of the buffered preamps he has designed.  

My personal preamp is a no-gain, buffered unit that was custom built by Steve McCormack and is very similar to his early VRE-1 preamp.  It works well with every amplifier I have used and sounds better than the over-20 highly regarded preamps that have paraded through my system over the years. Below is an excerpt from a 1995 interview with Steve where he was discussing his thoughts at the time on passives and buffering.  A link to the full article follows.

The sonic performance of the finished product, called the Line Drive, fulfilled all of the original design goals. It was characterized by the exceptional transparency and natural musicality which are now recognized as classic passive features when the preamplifier is properly designed and manufactured, yet rarely heard from other types of components.

With further experimentation, I discovered that the very best performance of my design came from running it with a high input impedance (20 kOhm or higher), high gain power amplifier. This too is characteristic of all passive units. The amplifier's high input impedance prevents loading effects while the high gain assures ample capability to drive loudspeakers to an appropriate volume level. Only extremely low loudspeaker sensitivity combined with low gain amplifiers present a problem for attaining high volume levels. Again, this combination of factors is rarely present in the real world.

While investigating early reports of "insufficient" volume, I found that people simply were not turning the volume control far enough to achieve the levels they wanted. Following the lessons they had learned about not turning up the volume of active preamplifiers too much as a precaution against system damage, these individuals were setting their controls too low. Since passive line-level preamplifiers have unity gain (1X), there is no reason not to turn the volume all the way up if that is required.

Please understand that I have nothing against active circuits, and I assume the same is true of other creators of passive designs. I build active units for applications where they are appropriate, such as phono preamplification. However, I prefer the passive path in some other applications because of the exceptional transparency and natural musicality possible through this approach.

During the latter years of the original Line Drive's seven year tenure, I worked with design refinements in preparation for releasing its successor. These experiments confirmed most of my earlier conclusions, but they also led me to appreciate the advantages of offering a buffered output in addition to the passive output. I had resisted that option for a while because the buffer circuits I knew had a negative effect on sonic performance. It was during the development of the Active Line Drive ALD-1 preamplifier that I designed a special buffer circuit that was based on a simple complimentary FET pair, but which required a very high quality power supply to make it work properly.

This was the first electronic circuit I had worked with that complimented the passive circuit and did not compromise sound quality. It allows the use of any length of cable between the preamplifier and power amplifier. Adding the buffered circuit brings the opportunity to enjoy the sonic benefits of minimalist preamplification to all systems.

Link:http://hometheaterhifi.com/volume_2_1/passive.html

The purpose of taking the time for this post is to provide information showing why I believe there is no one answer or product that is going to work well for everyone and to respond to the OP as to why simply taking things out of the signal path is not always a recipe for success.

Mitch2, thank you for taking the time to explain in detail the other variables involved and pointing out the genius behind Steve McCormacks designs which I believe are top notch!

Thanks for the detailed post Mitch.
I totally agree with you that if your source and amp gain and/or impedance are mismatched, an active unit may improve the overall result.
What i'm questioning is the common approach on how to solve this mismatch.
My suggestion is that you might be better off trying to find a better matched source-amp combination than introducing an active pre-amp to make them work well together. 
If you invest the money you save on the preamp, and get an upgraded and better matched source or amp, the overall result may be much better than having to introduce a preamp as a 'match maker' for a 'given' source and amp.

I would like to add a quote from Ted Smith that addresses cable capacitance as well, which was posted on the PS Audio forum regarding the DS DAC direct to amp approach. The key sentence here IMHO is

• "We don’t always build our systems from whole cloth where we might have the opportunity to find a set of components with no interface issues: a preamp is a good thing to have on hand for the cases where other factors like gain mismatch or cable length or… get in the way of a well balanced system." - 
  

So why not strive for a well balanced, preamp-less system to begin with?

Here is the full quote:

“There are two issues that come to mind: gain (which has already been mentioned) and the other is cable capacitance.

With respect to gain, there’s a “best” sensitivity of amp to use with the DS direct: you need enough headroom to have dynamic music (even on your louder tracks) but also enough sensitivity that you aren’t using the volume control far from 100 a lot of the time. Obviously if your music collection has tracks that are significantly different in loudness or dynamic range or if the amp isn’t sensitive enough or is way too sensitive you’ll probably want a preamp.
With cables that have too high of a capacitance there are FR response issues with almost any source. With most sources high capacitance implies a high frequency rolloff, but at times the transformer output of the DS can interact with cable capacitance to add a little high frequency boost. In an already existent system that already has high capacitance cables but is otherwise balanced changing from some other DAC to a DS may make quite a difference in the very top of the audio band. Either a rolloff or a boost of the highs could be beneficial in some setups, but in general, average to lower capacitance cables will be better with the DS or you’ll need a preamp to drive/buffer a higher capacitance or long run cables after the DS.
My counter argument to the minimalist point of view is that “Why should you expect a $6000 preamp in a $6000 DAC for free?” We don’t always build our systems from whole cloth where we might have the opportunity to find a set of components with no interface issues: a preamp is a good thing to have on hand for the cases where other factors like gain mismatch or cable length or… get in the way of a well balanced system.
Given a particular amp the DS could be designed so that you probably wouldn’t want a preamp – and similarly given the DS an amp could be designed so you probably don’t want a preamp, but in real life, as all things in audiophile land, you’ll need to listen for yourself to different setups and make up your own mind.”

If your source has sufficient output voltage, there are a couple of DACs that offer volume control solutions that maybe superior to even properly implemented passives, and may approach the sonic quality of very good active preamps.  These are implemented by changing the reference voltage of the dacs, which does not affect the "bits."

Steve Nugent's Empirical Audio Overdrive SE and SX offer this type of on-board volume control method as does Metrum's new Adagio.  I am currently enjoying Metrum's Pavane (which does not offer volume control) but I am very curious to try the Adagio, which is said to offer other sonic improvements over the already very good Pavane, in addition to the volume control option.

I have no clear idea how PS Audio's DirectStream DAC controls volume although it is said to be in the "digital domain."  It was surprising to me that Art Dudley didn't even try the direct-to-amp approach in  Stereophile's first review of the DirectStream DAC.  He brushed it off by saying;

Given the choice, I always prefer the sound of my system with an active preamplifier—the passive approach seems to me sorely lacking in drive by comparison—so I didn't try using the DirectStream to directly drive any of my amps.

Robert Deutsch did try it direct and had this to say in his 2015 follow-up;

Since the DS's analog output is variable, it can be used to drive an amplifier directly rather than running the signal through a preamp. Like most people who've compared a high-quality active preamp with a passive controller, I've found active preamps to be superior, particularly in dynamics. But the DS is different: its output level is controlled in the digital, not the analog domain, and Ted Smith, lead designer of the DS, claims that using the DS's variable output to drive an amp directly results in no loss of resolution. In fact, this is what PSA recommends for optimal sound quality.

Listening at matched levels through my Convergent Audio Technology SL-1 Renaissance preamp vs direct connection, I went back and forth at least a half-dozen times, trying to decide which was better—which should give you an idea of how close the sounds were. The sound through the CAT was a bit warmer, which was welcome with voices. The direct connection to the MC275LE resulted in even more finely defined detail, but perhaps veered slightly in the "clinical" direction. Overall, I preferred the connection through the preamp, but it was a close thing. The CAT is one of the best preamps around; pitted against a lesser preamp, the direct connection would likely be the winner.

He seemed to observe the "more detailed but thinner, dryer, less full, etc." perception that others report when running DACs direct to amplifiers, or when using passives, compared to using active preamps.  In fairness to PS Audio, AudioStream also preferred having an Ayre preamp in the signal path when compared to running the Empirical Audio Overdrive SE direct to their Ayre amplifiers.

Here is a part of the review from 6 Moons for the new Gryphon Antileon Evo poweramp, driving Wilson Alexia’s
What was used first as the preamp was his Supratek Reference DHT, and yes this quote of part of the review is pushing my product, but it goes to show at least to this reviewer what a passive can do.
And if his source (AMR CD-77.1) had a volume control of it’s own, it would have sounded even better to him direct.

That’s the quintessential description for the sound of the Antileon EVO, too. There’s a sense of weight, power, dominance and uncompromising authority to music when played via this amplifier. And it’s not just about the bass either. It’s about overall dynamics that approach reality (yes, rest of gear permitting…) with explosive contrasts. This last came through even more outwardly when I connected George Stantscheff’s Lightspeed Attenuator quad-matched LDR passive. The Lightspeed does no harm. It’s as clear a conduit to the musical content from your source as you’re going to get, provided all impedance parameters are optimal, and without hindering dynamic expression in any way, shape or form.

Cheers George