"What HiFi" - can it be trusted?
 

"Ian Molton" Wrote

Exactly my point. So why have a system that highlights a bad recording
thereby distracting one from the performance?

It ain't rocket science.

If I went to a bad performance I'd walk out, not put in earplugs so I
couldnt hear the awful treble racket from the 3rd violin.

Just out of interest..... How many bad performances have you walked out of?
And which was the worst? I can only remember walking out of 1, from a pub
in Killin, where the "entertainment" was truly, spectacularly bad. I've
walked out of a couple of films But it's incredibly hard to do!! I always
feel I should appologise "really sorry, can't stay, can't stand this, this
is crap, sorry, sorry, sorry" and thats just to a film!

David

"What HiFi" - can it be trusted?
 

"Ian Molton" Wrote

Exactly my point. So why have a system that highlights a bad recording
thereby distracting one from the performance?

It ain't rocket science.

If I went to a bad performance I'd walk out, not put in earplugs so I
couldnt hear the awful treble racket from the 3rd violin.

Just out of interest..... How many bad performances have you walked out of?
And which was the worst? I can only remember walking out of 1, from a pub
in Killin, where the "entertainment" was truly, spectacularly bad. I've
walked out of a couple of films But it's incredibly hard to do!! I always
feel I should appologise "really sorry, can't stay, can't stand this, this
is crap, sorry, sorry, sorry" and thats just to a film!

David

"What HiFi" - can it be trusted?
 

On Mon, 12 Jan 2004 18:10:03 +0000 (UTC)
"David" wrote:


Just out of interest..... How many bad performances have you walked
out of?

Ive only been near two I would have walked out of. sadly as I was
employed at the establishment, all I could do was appologise to the
custiomers...

--
Spyros lair: http://www.mnementh.co.uk/ |||| Maintainer: arm26 linux

Do not meddle in the affairs of Dragons, for you are tasty and good with
ketchup.

"What HiFi" - can it be trusted?
 

On Mon, 12 Jan 2004 18:10:03 +0000 (UTC)
"David" wrote:


Just out of interest..... How many bad performances have you walked
out of?

Ive only been near two I would have walked out of. sadly as I was
employed at the establishment, all I could do was appologise to the
custiomers...

--
Spyros lair: http://www.mnementh.co.uk/ |||| Maintainer: arm26 linux

Do not meddle in the affairs of Dragons, for you are tasty and good with
ketchup.

"What HiFi" - can it be trusted?
 

On Mon, 12 Jan 2004 17:29:37 +0000
Glenn Booth wrote:

Hi,

In message , Ian Molton
writes

perfect inks and a perfect monitor will not have any colours that
each cannot reproduce (as far as the eye is concerned).

Yes they will, because there are simply less possible values in the
CMYK colour gamut than there are in the RGB gamut,

That is just not true. and the very fact you say CMYK implies that you
are dealing with imperfect CMY components that require the K to enhance
them where they fall short.

RGB and CMY are merely different views on the same colourspace.

and the eye has a larger colour space than either.

Poppycock. Unless you are limiting your CMY values to a fixed number of 'bits' each.

Even if perfect inks existed, the colours
aren't defined. You can create an image that fits into the CMYK space
*and* the RGB space, but you can't ever guarantee that a 'legal and
valid' RGB image can be converted to CMYK without loss of information.

given the limits of inks (which is what you deal with in publishing/design) yes, thats true.

Since graphics cards running in true colour don't use look up
tables,

No, thats Directcolour. Truecolour has the same number of colours on
the output and *DOES* have 8-bit lookup tables, one per colour
component. Gamma (in truecolour modes) on many cards is programmed by

writing an appropriate table to the LUT.

Are you thinking of palletised colour systems, like the 256 colour
mode in Windows?

No, Im not. Read up, or google the term.

*directcolour* is where the bits in the pixel feed straight into the output DACs.
*true* colour is where the bits feed into LUTs first.

The graphics
ASICs I know best are the ones made by my employer (Matrox) and they
use 10 bit DACs that run up to 400MHz, with no look up tables in 16,
24, 30 or 32 bits per pixel. Gamma tables however, are stored in
programmable 10 bit LUTs.

if they are purely GAMMA they wont be 'tables' in a normal sense. however they may well just be general putpose LUTs - are you SURE you can only program them with gamma values? are then not in fact directly addressable ?

besides, once you get into nonlinearity of the monitor etc. the whole
idea of a simple curve like gamma is silly. Then you need to add the
paper into the equation as well...

And the phosphors, and the ink batch, and the age of the light bulbs
that are overhead. That's why I said it's a bag of worms. I guess I
get a bit overexcited about colour science, sad git that I am.

I guess you may hae noticed this is an area I've had more than a passing interest in ;-) (I was on the verge of completely re-writing the linux voodoo3 drivers at one point but other commitments sunk that project sadly...)

--
Spyros lair: http://www.mnementh.co.uk/ |||| Maintainer: arm26 linux

Do not meddle in the affairs of Dragons, for you are tasty and good with
ketchup.

"What HiFi" - can it be trusted?
 

On Mon, 12 Jan 2004 17:29:37 +0000
Glenn Booth wrote:

Hi,

In message , Ian Molton
writes

perfect inks and a perfect monitor will not have any colours that
each cannot reproduce (as far as the eye is concerned).

Yes they will, because there are simply less possible values in the
CMYK colour gamut than there are in the RGB gamut,

That is just not true. and the very fact you say CMYK implies that you
are dealing with imperfect CMY components that require the K to enhance
them where they fall short.

RGB and CMY are merely different views on the same colourspace.

and the eye has a larger colour space than either.

Poppycock. Unless you are limiting your CMY values to a fixed number of 'bits' each.

Even if perfect inks existed, the colours
aren't defined. You can create an image that fits into the CMYK space
*and* the RGB space, but you can't ever guarantee that a 'legal and
valid' RGB image can be converted to CMYK without loss of information.

given the limits of inks (which is what you deal with in publishing/design) yes, thats true.

Since graphics cards running in true colour don't use look up
tables,

No, thats Directcolour. Truecolour has the same number of colours on
the output and *DOES* have 8-bit lookup tables, one per colour
component. Gamma (in truecolour modes) on many cards is programmed by

writing an appropriate table to the LUT.

Are you thinking of palletised colour systems, like the 256 colour
mode in Windows?

No, Im not. Read up, or google the term.

*directcolour* is where the bits in the pixel feed straight into the output DACs.
*true* colour is where the bits feed into LUTs first.

The graphics
ASICs I know best are the ones made by my employer (Matrox) and they
use 10 bit DACs that run up to 400MHz, with no look up tables in 16,
24, 30 or 32 bits per pixel. Gamma tables however, are stored in
programmable 10 bit LUTs.

if they are purely GAMMA they wont be 'tables' in a normal sense. however they may well just be general putpose LUTs - are you SURE you can only program them with gamma values? are then not in fact directly addressable ?

besides, once you get into nonlinearity of the monitor etc. the whole
idea of a simple curve like gamma is silly. Then you need to add the
paper into the equation as well...

And the phosphors, and the ink batch, and the age of the light bulbs
that are overhead. That's why I said it's a bag of worms. I guess I
get a bit overexcited about colour science, sad git that I am.

I guess you may hae noticed this is an area I've had more than a passing interest in ;-) (I was on the verge of completely re-writing the linux voodoo3 drivers at one point but other commitments sunk that project sadly...)

--
Spyros lair: http://www.mnementh.co.uk/ |||| Maintainer: arm26 linux

Do not meddle in the affairs of Dragons, for you are tasty and good with
ketchup.

"What HiFi" - can it be trusted?
 

On Mon, 12 Jan 2004 14:04:01 -0000, "Keith G"
wrote:


"Stewart Pinkerton" wrote in message
...
On Sun, 11 Jan 2004 22:40:44 -0000, "Keith G"
wrote:

"Jim Lesurf" wrote in message
...

2) SS amps - May be more accurate, so the amp has relatively little
'sound'
of its own, hence the 'sound' depends more on the input than the amp.
Some
people prefer this as it allows them to hear more clearly what was
recorded
or broadcast and avoids applying the same 'effect' to everything they
hear.

OK, but very often has other 'effects' like killing the imaging, timbre
and
detail as well as trapping the sound firmly in the same plane as the
speakers.....

No Keith, those things were never in the input signal. What you like
about valves is the artifacts that they *add* to the sound, not
anything that SS amps magically remove.

Lack - I never said 'remove'.....

They 'lack' those things because they were *not* in the input signal.

BTW, SS amps certainly do not
'trap the sound firmly in the same plane as the speakers' if you have
decent speakers.

You tell me - Ruark Paladins and B&W CDM1SEs (all 'rock steady') atm.....

See my paragraph below - they qualify!

The vague phasiness of a lot of valve amps might help
out some typically cold and clinical modern box speakers, but that's
still an artifice, not *high fidelity* sound.


'HiFi' is as 'HiFi' does - means a lot of different things to different
people. I dislike the phrase other than to distinguish 'home audio
entertainment' kit from, say, 'home theatre' gear'. The phrase 'the stereo'
will do equally well, often as not.

My requirements differ from those of the normal 'audiophile', I think - my
valves and (you know what) sound far more 'lifelike' to me than CD/SS ever
could. (Tried enough of it, both ways, in my time.)

Fair enough, there's no arguing with preference.

If it helps, I readily admit a strong preference for 'listenability' over
'metered accuracy' but I ain't stupid enough to know that 'accuracy' isn't
important - it's a question of to what degree and at what cost AFAIAC......

Can't argue with that - there are *no* perfect speakers - not even
close, not even at £50k a pop! :-(
--

Stewart Pinkerton | Music is Art - Audio is Engineering

"What HiFi" - can it be trusted?
 

On Mon, 12 Jan 2004 14:04:01 -0000, "Keith G"
wrote:


"Stewart Pinkerton" wrote in message
...
On Sun, 11 Jan 2004 22:40:44 -0000, "Keith G"
wrote:

"Jim Lesurf" wrote in message
...

2) SS amps - May be more accurate, so the amp has relatively little
'sound'
of its own, hence the 'sound' depends more on the input than the amp.
Some
people prefer this as it allows them to hear more clearly what was
recorded
or broadcast and avoids applying the same 'effect' to everything they
hear.

OK, but very often has other 'effects' like killing the imaging, timbre
and
detail as well as trapping the sound firmly in the same plane as the
speakers.....

No Keith, those things were never in the input signal. What you like
about valves is the artifacts that they *add* to the sound, not
anything that SS amps magically remove.

Lack - I never said 'remove'.....

They 'lack' those things because they were *not* in the input signal.

BTW, SS amps certainly do not
'trap the sound firmly in the same plane as the speakers' if you have
decent speakers.

You tell me - Ruark Paladins and B&W CDM1SEs (all 'rock steady') atm.....

See my paragraph below - they qualify!

The vague phasiness of a lot of valve amps might help
out some typically cold and clinical modern box speakers, but that's
still an artifice, not *high fidelity* sound.


'HiFi' is as 'HiFi' does - means a lot of different things to different
people. I dislike the phrase other than to distinguish 'home audio
entertainment' kit from, say, 'home theatre' gear'. The phrase 'the stereo'
will do equally well, often as not.

My requirements differ from those of the normal 'audiophile', I think - my
valves and (you know what) sound far more 'lifelike' to me than CD/SS ever
could. (Tried enough of it, both ways, in my time.)

Fair enough, there's no arguing with preference.

If it helps, I readily admit a strong preference for 'listenability' over
'metered accuracy' but I ain't stupid enough to know that 'accuracy' isn't
important - it's a question of to what degree and at what cost AFAIAC......

Can't argue with that - there are *no* perfect speakers - not even
close, not even at £50k a pop! :-(
--

Stewart Pinkerton | Music is Art - Audio is Engineering

"What HiFi" - can it be trusted?
 

On Mon, 12 Jan 2004 14:09:00 -0000, "Keith G"
wrote:

Cheapest tweak in the world is get out of the chair and move back a bit or
even out of the room if necessary. (Says that sitting in a bungalow....)

Hmmmm, I wonder if Kurt lives in a tower block............... :-)
--

Stewart Pinkerton | Music is Art - Audio is Engineering

"What HiFi" - can it be trusted?
 

On Mon, 12 Jan 2004 14:09:00 -0000, "Keith G"
wrote:

Cheapest tweak in the world is get out of the chair and move back a bit or
even out of the room if necessary. (Says that sitting in a bungalow....)

Hmmmm, I wonder if Kurt lives in a tower block............... :-)
--

Stewart Pinkerton | Music is Art - Audio is Engineering