1. Home
  2. analog
  3. 198755-1647

What does "Warm" mean?

In his post today Paul McGowan ([email protected]) had a pretty good post on the subject of warmth.  Here is his definition: "The word usually points at a few related things. There's a slight rise in the lower midrange and upper bass — the region where male voices, cellos, lower piano notes, and the body of most instruments live. There's a softness at the very top, a rounding off of transient energy that feels easier on the ear over long sessions. And there's a sense of weight and body in the music, a presence that makes acoustic instruments feel three-dimensional rather than etched in air. When all three of those things show up together, listeners reach for "warm" because nothing else describes it as quickly."

Paul then goes on to discuss both the good and the bad of this set of characteristics.  In general I associate warmth with harmonic distortion, but prefer it to the stridency or harshness that I associate with intermodular distortion.  I think most people would agree.  Since we all live with a certain amount of distortion, more in analog than in digital playback, I wonder what others think about warmth, or lack of same, and their tolerance or lack of same in their own listening preferences. 

billstevenson
atmasphere's avatar
atmasphere
12,430 posts
 

@billstevenson 'Warmth' can be caused by something other than a frequency response error.

We know this because certain amplifiers which have bandwidth well past 50KHz can sound 'warm'.

They do it by generating lower ordered harmonics; the 2nd and 3rd. These harmonics are prodigious enough they are able to mask higher orders. When this quality is present in an amplifier or preamp, it will sound 'warm' even though its bandwidth is wide and its frequency response is flat.

SETs are a good example of what I'm talking about. Many of them are not capable of full power at 20Hz or even 40Hz; despite this low frequency roll off they still sound 'warm'. 

The ear has a built in tone control. If bass is lacking, the system will sound tilted towards the highs. If there is too much bass, the highs will sound muffled. Thus SETs clearly demonstrate that frequency response isn't the primary cause of 'warmth' while distortion is.

faustuss's avatar
faustuss
890 posts
 

@atmasphere 

"SETs are a good example of what I'm talking about. Many of them are not capable of full power at 20Hz or even 40Hz; despite this low frequency roll off they still sound 'warm'."

There may actually be very little roll off at low frequencies at all until you attempt to meet or exceed it's 1Khz power rating wherein bass distortion will rise exponentially and exceeding the amps 1Khz distortion rating. The Mfg. probably doesn't honestly state distortion and power bandwidth for rated power as the FCC requires for solid state so its anybody's guess. As you once said their specs are "arbitrary" and the power rating is quite "arbitrary" and farcical to begin with. The warmth can be associated with excessive amounts of intermodulation distortion not harmonic.

atmasphere's avatar
atmasphere
12,430 posts
 

There may actually be very little roll off at low frequencies at all until you attempt to meet or exceed it's 1Khz power rating

@faustuss IMD is interpreted by the ear as harshness as are higher ordered harmonics. Lower ordered harmonics are not.

Most SETs cannot make full power at lower frequencies (less than 40Hz; many are rolling off an octave higher) since the output transformer is in series with the power tube, requiring a gap in the core to prevent DC saturation. This causes a loss of inductance at low frequencies, which is the cause of the rolloff, since the load on the power tube at those frequencies is not only lower, but also elliptical rather than linear (hence higher distortion as well as lower power). 

More to discover