Why some preamps does not play music? Cold?

P.S. Guys, this is pure technical question. I'm not asking which preamp does it all, so please no preamp recommendations!!! I just want to understand what is causing said phenomena, that is it.
Thank you for understanding.

The reason a preamp may or may not be emotionally involving has usually to do with distortion. I'm not talking about a lot here either. Just a small amount of higher ordered harmonic distortion that is difficult to measure is easily heard by the human ear as brightness.

This is because the human ear/brain system converts all forms of distortion into some sort of tonality. Higher ordered harmonics are an excellent example- and have a further complication with the human ear as the ear uses those harmonics to gauge how loud a sound is. So when they get messed with, the ear hears it as louder, harsher and brighter.

Music is processed in the limbic centers of the brain. However, if the brain detects that there is something wrong somewhere, it has a tipping point wherein the music processing is transferred from the limbic centers to the cerebral cortex (where consciousness/intellect resides).

When this happens, the emotional connection is lost. No more foot tapping, none of that feeling of wanting to dance.

So what the designer of the electronics has to be aware of is what forms of distortion to which the brain is most sensitive, and avoid them through design. This is not really that hard if you understand engineering, what is hard is that you have to know the engineering and the physiology at the same time.

One example of a design technique that causes a loss of emotional impact is the use of loop feedback in the circuit. Loop feedback, while overall suppressing distortion to a great degree, actually **adds** higher ordered harmonics to the signal. You already know the effect of that. But it does more- there are usually intermodulations at the point that the feedback is applied (the feedback node). These intermodulations may not have any musical relationship to the signal at all, at being at a low level, exist more as part of the noise floor of the preamp or amp. This has been known for the better part of 60 years (see Norman Crowhurst- he was writing about this in the 1950s).

Now it happens that the ear has a masking principle- wherein the presence of a louder sound will mask the presence of a quieter sound (this is the basis of the encoding of mp3 files BTW). This is well-known. What is less well known is that there is an exception to the masking rule that has to do with hiss. This is likely evolutionary as wind and water make hissing sounds in the natural environment, and the inability to hear noises below the hiss might have profound survival aspects!

So we can hear about 10-20 db into a natural hiss noise floor (the exact amount being a matter of the individual and also debate, but the 10db figure seems to be a solid). Now if the noise floor of the circuit is composed of harmonic and inharmonic (intermodulation) noise (as opposed to the hiss that arises from noise sources in the circuit unrelated to loop feedback), the result is that even though consciously we can't hear a lot of difference between the two noise floors, the unconscious portion of our brain (which is about 95% of the brain structure) can detect that something is wrong. 

When it does so, music processing is transfered to the cerebral cortex: the emotion connection is lost.

Does that help with the understanding? Any questions?

However, as it most apparent on complex, avangarde jazz pieces (Kith Jarett, Jan Garbarek, Chick Corea etc.) it looses musical connection (?) It does it on any kind of music, but on regular music it is less obvious, because it still have enough "melody (?)" left to connect sounds into music(?). When switched to any other, out of 5 preamps, (cold, SS or whatever) which BTW, have less of extensions, clarity etc.,, the music comes back. So, the correct word is probably musicality not emotional engagement?

Usually tubes have lower distortion (after all, they are quite linear). However, a lot depends on execution- and the tubes in the unit. You might have them tested for starters; a bad tube might do OK on simple material and might fall apart when things get complex.

Other brands might work better too.

Also there is the design of the line section itself- they are not all created equal and some are pretty bad- as George points out, they can be prone to colorations of their own (usually the distortions found in tube equipment are of the lower harmonics, which the ear converts to 'warmth'). But if there are design flaws all sorts of issues can crop up.

I would start by getting the tubes tested!

However, it is 6sn7GTB therefore not to much choice to play with, unless try not GTB but that, as i understand, not recommended

There are plenty of 6SN7 types out there, and in a preamp I would expect that a GT, GTA or GTB could be used. You might want to check with the manufacturer on that. The big difference with the suffixes is how much voltage the tube can handle.

6SN7s, FWIW, are an excellent tube (the only bad ones being the modern Russian types) and so I would not expect that tube to be the issue, unless its microphonic or something like that.

as a matter of fact, those Sylvania which i have currently installed are extremely microphonic. Are you suggesting it might cause the issue i'm talking about? 

Microphonics can cause brightness and amusicality. Any tube used for audio should be hand-picked for low microphonics.