Goldpoint SA4M-47 passive 'preamp' winging its way to Marco Towers...

[awkwardbydesign]

I was recently reading Valve Amplifiers Third Edition by Morgan Jones again and his thoughts on LDR

"

"

Well, that's interesting. 0.02% of 2nd harmonic may actually sound better, depending on your mental map! And the poor matching is well known, which is why they have to be matched. Drift can also be a problem, but experiences vary. As for unmeasurable distortion with switched attenuators, well just ask Marco! Inaudible would be a better aim.
This is a problem with accepting measured results, it depends on what is measured, how accurate the measurement, how it is interpreted, and so on. We aren't testing life saving drugs here, so subjective experiences are just as valid as measured data, where listening to music is concerned.

[lurcher]

Morgan's text is a little misleading though, as firstly he is not using the LDRs that are now used in preamps, secondly he is only using one, when most designs use two, and thirdly he assumed that is the distortion was high at 10v, low at 2v it would be lower still at lower voltage. The first point means the measured level of distortion is not what you see in the current units, the second point means the distortion is 3rd Harmonic, not 2nd, and the third point means that the transfer curve of the stage varies with volume setting and signal level.

Of course, if they sound better to you thats great everyone wins, but dont assume they are sounding better because they are more transparent.

[anubisgrau]

so it's a resistor, not a transistor

[lurcher]

so it's a resistor, not a transistor

No its a semiconductor that has a resistance that varies with light level. No one said it was a transistor.

[awkwardbydesign]

Of course, if they sound better to you thats great everyone wins, but dont assume they are sounding better because they are more transparent.

I'm amused by the use of visual analogies. I used to work in stained glass, and ruby glass (among many others) is transparent! Not uncoloured, just not opaque or translucent. So calling something audibly transparent doesn't mean what most people think it does. IIRC, Linn Kans were "transparent", which helps* to explain their popularity, but they sure as hell weren't uncoloured! I have no objection to the term as such, just the misuse and misunderstanding of it.
* Linn marketing being at least as big a reason. But I heard them at the time, and could understand the appeal, especially in comparison to the many thick** sounding but good measuring alternatives.
** Translucent?

[lurcher]

I used to work in stained glass, and ruby glass (among many others) is transparent!

Not below 500nm its not.

I could have used more accurate language, I could have said it did not have a linear transfer function, or I could have shown a FFT of its distortion spectra, but given the nature of most readers of this forum I chose language that I believed would make sense to the most people.

[Marco]

Well, that's interesting. 0.02% of 2nd harmonic may actually sound better, depending on your mental map! And the poor matching is well known, which is why they have to be matched. Drift can also be a problem, but experiences vary. As for unmeasurable distortion with switched attenuators, well just ask Marco! Inaudible would be a better aim.
This is a problem with accepting measured results, it depends on what is measured, how accurate the measurement, how it is interpreted, and so on. We aren't testing life saving drugs here, so subjective experiences are just as valid as measured data, where listening to music is concerned.

I wholeheartedly concur (on all points), but specifically the bit in blue!

That bit should be displayed on the banner of every hi-fi forum in the land, in big, bold flashing neon lights!!

Marco.

[Macca]

Whoever claimed Linn Kans were transparent?

Transparent has a pretty well established meaning in audio which is a component/process that adds nothing measurable to the signal. So no speaker - not even a Linn Kan - can be transparent.

You will see people using it in other contexts i.e 'transparent sounding' when what they actually mean is 'doesn't sound all horrid and distorted' or 'I can hear what the vocalist is singing' but this isn't transparancy it just means the system being described is not total crap.

[DSJR]

ALL mk1 and 2 Kans are horrid and distorted - it was designed into them and it got worse as KEF messed up the bass-mid units. The mk2 was a serious attempt to make them more neutral, but the box colouration became unbearable to me - ugh! Having that whacking peak up at very upper midrange made drum-kits sound impressive though

Anyway, do resistor networks add distortion? Surely a proper attenuator need only have two value-selected resistors in circuit at any one setting. That way the loading as seen by source and amp would be the same, thereby removing yet another variable, would it not?

This whole thread seems to me like a hi-tech sledgehammer trying to crack a peanut, when a little plastic-film pot would do the trick? Sorry, I just don't 'get' the 'have to improve what the designer thought was fine' vibe - not criticising, just increasingly confused.

[Joe]

Whoever claimed Linn Kans were transparent?

Transparent has a pretty well established meaning in audio which is a component/process that adds nothing measurable to the signal. So no speaker - not even a Linn Kan - can be transparent.

You will see people using it in other contexts i.e 'transparent sounding' when what they actually mean is 'doesn't sound all horrid and distorted' or 'I can hear what the vocalist is singing' but this isn't transparancy it just means the system being described is not total crap.

Valves are more transparent than transistors because they're made of glass and you can see through them.