Hello and some A1 questions about exposure compensation, flash and RAW

Hello everybody

I've recently spoted a great deal on eBay of a second hand A1, so I decided to let go of my Canon S50 and get a real camera

I have previously chosen the S50 because of the manual controls on my price range, with the intention of learning more about photography, but altough it had most manual options, they were a pain to use, so I ended up using it just a Point&Shoot camera.

Now with the A1 it's a different story. There are buttons for everything No need to go to the menus. So I'm finally starting to learn something about Photography.

I've been reading a lot on these forums and on other sites (like luminous landscape, dpFWIW, etc...), but I still have many doubts.

This will be a first of some posts asking for help on understanding how things work.

For now I'm puzzled about some things in RAW mode, mainly exposure compensation.

1) I thought that the only thing that exposure compensasion did was adjust Shutter Speed and Aperture in ways that differed from the auto metering of the camera. Hence no exposure compensation on manual mode.

But when I open my RAW files in DiMAGE Viewer I have the option of changing it. How can that be? Surely the shutter and aperture can't be changed now So what's the converter really doing?

2) And while on exposure compensation, what does the flash exposure compensation really does? Modify the intensity of the flash, act as an independent exposure compensation setting that is used when the flash is used, or a combination of both? I'm inclined to think it is the first, as in flash manual mode the setting is gone, but i want to make sure it'ss not the third.

3) If we shoot RAW, the Sharpness, Contrast, Color and Filter options don't have any effect, right? Just specify a "default". So what about the color mode? Does it makes a difference to specify it on RAW? I was thinking we could have a change to specify the color mode when we do the conversion from RAW to a standard format like JPEG, TIFF or PNG, but I don't see that option. Also, are the B&W and Solarization different ways to convert to sRGB, like vivid?

Thanks in advance
Gustavo Guerra

ovatsus said:

Hello everybody

I've recently spoted a great deal on eBay of a second hand A1, so I
decided to let go of my Canon S50 and get a real camera

I have previously chosen the S50 because of the manual controls on
my price range, with the intention of learning more about
photography, but altough it had most manual options, they were a
pain to use, so I ended up using it just a Point&Shoot camera.

Now with the A1 it's a different story. There are buttons for
everything No need to go to the menus. So I'm finally starting
to learn something about Photography.

I've been reading a lot on these forums and on other sites (like
luminous landscape, dpFWIW, etc...), but I still have many doubts.

This will be a first of some posts asking for help on understanding
how things work.

For now I'm puzzled about some things in RAW mode, mainly exposure
compensation.

1) I thought that the only thing that exposure compensasion did was
adjust Shutter Speed and Aperture in ways that differed from the
auto metering of the camera. Hence no exposure compensation on
manual mode.
But when I open my RAW files in DiMAGE Viewer I have the option of
changing it. How can that be? Surely the shutter and aperture can't
be changed now So what's the converter really doing?

2) And while on exposure compensation, what does the flash exposure
compensation really does? Modify the intensity of the flash, act as
an independent exposure compensation setting that is used when the
flash is used, or a combination of both? I'm inclined to think it
is the first, as in flash manual mode the setting is gone, but i
want to make sure it'ss not the third.

3) If we shoot RAW, the Sharpness, Contrast, Color and Filter
options don't have any effect, right? Just specify a "default". So
what about the color mode? Does it makes a difference to specify it
on RAW? I was thinking we could have a change to specify the color
mode when we do the conversion from RAW to a standard format like
JPEG, TIFF or PNG, but I don't see that option. Also, are the B&W
and Solarization different ways to convert to sRGB, like vivid?

Thanks in advance
Gustavo Guerra

Click to expand...

I'm not a expert in the tecnical side of raw shooting allthrough I only
shoot in raw. So please straight my mistakes up out there.
1) Shure you can't change it, but you can change nearly to what it
would had been.
2) Flash compensation is very usefull with fill-flash and portraits, to get
deep shadows away, but not too much.
3) Some of these items may work in raw, but it is better to do all
compensation in the raw-converter, in your photo-editor or in
special programs, best in 3 x 16 bit color.
I hope you got some answers.
http://jens.noehr.person.emu.dk/
--
Jens Noer

Hi Gustavo,

Congratulations on your A1, I have been using mine for several months now and have been very happy with the results.

To add on Jens comments with regard exposure compensation. When you take the shot, exposure compensation will lighten or darken the shot by modifying the shutter speed or aperture. For example, if you are in aperture priority, setting a negative EV, will give a faster shutter speed. Once the shot is taken, the exposure compenation in the raw converter simply lightens or darkens the recorded shot.

The advantage of doing this in the raw converter is that the raw format has not been processed through jpeg conversion. The jpeg format compresses an image by using using a single data point to represent multple similar adjacent data points. For example, if your shot had an area of very dark shadow, jpeg might translate this as a area of pure black, where there might be some subtle variation of detail that would still be represented in the raw version of the image.

I can't comment on flash compensation, as I very seldom flash.

With regard to color modes. I believe that choosing vivid or natural is relflected in the original raw image. I have a strong personal preference for natural color, so I simply don't use vivid color. Its easy enough increase saturation in PP, if needed.

Hope this helps,
Richard.

ovatsus said:

Hello everybody

I've recently spoted a great deal on eBay of a second hand A1, so I
decided to let go of my Canon S50 and get a real camera

I have previously chosen the S50 because of the manual controls on
my price range, with the intention of learning more about
photography, but altough it had most manual options, they were a
pain to use, so I ended up using it just a Point&Shoot camera.

Now with the A1 it's a different story. There are buttons for
everything No need to go to the menus. So I'm finally starting
to learn something about Photography.

I've been reading a lot on these forums and on other sites (like
luminous landscape, dpFWIW, etc...), but I still have many doubts.

This will be a first of some posts asking for help on understanding
how things work.

For now I'm puzzled about some things in RAW mode, mainly exposure
compensation.

1) I thought that the only thing that exposure compensasion did was
adjust Shutter Speed and Aperture in ways that differed from the
auto metering of the camera. Hence no exposure compensation on
manual mode.
But when I open my RAW files in DiMAGE Viewer I have the option of
changing it. How can that be? Surely the shutter and aperture can't
be changed now So what's the converter really doing?

2) And while on exposure compensation, what does the flash exposure
compensation really does? Modify the intensity of the flash, act as
an independent exposure compensation setting that is used when the
flash is used, or a combination of both? I'm inclined to think it
is the first, as in flash manual mode the setting is gone, but i
want to make sure it'ss not the third.

3) If we shoot RAW, the Sharpness, Contrast, Color and Filter
options don't have any effect, right? Just specify a "default". So
what about the color mode? Does it makes a difference to specify it
on RAW? I was thinking we could have a change to specify the color
mode when we do the conversion from RAW to a standard format like
JPEG, TIFF or PNG, but I don't see that option. Also, are the B&W
and Solarization different ways to convert to sRGB, like vivid?

Thanks in advance
Gustavo Guerra

Click to expand...

--
Richard B.
http://www.pbase.com/richard_b
http://www.rbpics.com

I'll try to answer questions one and three and leave the flash question to someone else.

1. The raw file is not actually a color image but rather a grey-scale reflecting the intensity of the electric output of each pixel as determined by the amount of light striking it. These values have been converted by the camera's A/D converter to a 12 bit series containing 4,096 levels. When the raw is converted to a color (RGB) image the values for each color channel are either written in 8 bits (256 levels) or 16 bits (65,536 levels). Clearly the values need to be reduced to fit into 8 bits or increased to fill the 16 bit space. However, if the change is made by a consistent (linear) coefficient, because of the nature of human vision, the result will look flat. Therefore, the conversion is done along a contrast curve which gives greater weight to the higher values. The shape of this curve can be varied, pushing values toward the right side of the histogram and thus "increasing exposure" or the opposite, decreasing "exposure" Two caveats - overdoing it can cause clipping (values pushed to the limits of the range and therefore devoid of details) and increased exposure comes at the price of also increasing the visibility of noise in the darker areas, so it is in general better to try to get the exposure right in the camera. Nevertheless, it can be a useful technique in situations which force you to underexpose, such as low light or high shutter speed to freeze motion.

2. Raw being a grey-scale file is not in any "color" space. However just as the camera attaches notes to the file which inform Divu what the camera settings were for contrast, sharpness, etc., so there is one for color space also. First however Divu translates the data to a color space appropriate for the particular camera model, called for instance Dimage A1-rprof.icc, then if in the Color Management dialog you have chosen "Original Space" it converts to the camera setting. If you have chosen Vivid it also pushes the saturation up. B/W means a black and white jpg and solarisation changes (falsifies) the colors of a jpg. Both of these can be done much better in your computer.
Elie

EliezerDinur said:

I'll try to answer questions one and three and leave the flash
question to someone else.
1. The raw file is not actually a color image but rather a
grey-scale reflecting the intensity of the electric output of each
pixel as determined by the amount of light striking it. These
values have been converted by the camera's A/D converter to a 12
bit series containing 4,096 levels. When the raw is converted to a
color (RGB) image the values for each color channel are either
written in 8 bits (256 levels) or 16 bits (65,536 levels). Clearly
the values need to be reduced to fit into 8 bits or increased to
fill the 16 bit space. However, if the change is made by a
consistent (linear) coefficient, because of the nature of human
vision, the result will look flat. Therefore, the conversion is
done along a contrast curve which gives greater weight to the
higher values. The shape of this curve can be varied, pushing
values toward the right side of the histogram and thus "increasing
exposure" or the opposite, decreasing "exposure" Two caveats -
overdoing it can cause clipping (values pushed to the limits of the
range and therefore devoid of details) and increased exposure comes
at the price of also increasing the visibility of noise in the
darker areas, so it is in general better to try to get the exposure
right in the camera. Nevertheless, it can be a useful technique in
situations which force you to underexpose, such as low light or
high shutter speed to freeze motion.

Click to expand...

Humm, so changing the exposure compensation in the RAW converter is the same as adjusting the levels after the conversion, if we use a lossless adn 16bit final format, right?

EliezerDinur said:

2. Raw being a grey-scale file is not in any "color" space. However
just as the camera attaches notes to the file which inform Divu
what the camera settings were for contrast, sharpness, etc., so
there is one for color space also. First however Divu translates
the data to a color space appropriate for the particular camera
model, called for instance Dimage A1-rprof.icc, then if in the
Color Management dialog you have chosen "Original Space" it
converts to the camera setting. If you have chosen Vivid it also
pushes the saturation up. B/W means a black and white jpg and
solarisation changes (falsifies) the colors of a jpg. Both of these
can be done much better in your computer.
Elie

Click to expand...

So, the converter renders the image with the Dimags A1-rprof.icc, and then converts to Color Space specified in options and then if we also have a Monitor Profile, it converts again. Ok, I think I got it. One thing I don't get is why there is a Original Color Space(sRGB) and a Original Color Space(AdobeRGB), if it's the Original, specified in the camera... Weird
In the Adobe Raw Converter than option doesn't even appear.

I tried taking the same foto in Natural, Vivid and Solarization. In bot DiVu and ACR, I didn't notice any difference. Neither in previewing them in ACDSee. So they must be ignored... Only in the thumbnail view in DiVu the solarized image appeared different.
Is my line of though right?

So, I can just stick with any setting in RAW mode, it won't make difference. Good, less things to worry about

Thanks for you feedback

ovatsus said:

EliezerDinur said:

I'll try to answer questions one and three and leave the flash
question to someone else.
1. The raw file is not actually a color image but rather a
grey-scale reflecting the intensity of the electric output of each
pixel as determined by the amount of light striking it. These
values have been converted by the camera's A/D converter to a 12
bit series containing 4,096 levels.

Click to expand...

Click to expand...

Humm, I've double checked the specs, and the A1 has 14 bits. The A2 only has 12bits, and I've noticed some posts mention 12 bits for both A1 and A1. O, is the A1 really 14bits or 12bits? I find it odd that the A2 has less color bit range than the A1.

ovatsus said:

ovatsus said:

EliezerDinur said:

I'll try to answer questions one and three and leave the flash
question to someone else.
1. The raw file is not actually a color image but rather a
grey-scale reflecting the intensity of the electric output of each
pixel as determined by the amount of light striking it. These
values have been converted by the camera's A/D converter to a 12
bit series containing 4,096 levels.

Click to expand...

Click to expand...

Humm, I've double checked the specs, and the A1 has 14 bits. The A2
only has 12bits, and I've noticed some posts mention 12 bits for
both A1 and A1. O, is the A1 really 14bits or 12bits? I find it odd
that the A2 has less color bit range than the A1.

Click to expand...

I agree with what your other correspondents wrote.

Well done for spotting the 12/14bit difference on the A1 / A2.

I have a background in electronic signal processing, and after reading the specs on the A1, I also assumed that the A2 would have 14bit colour depth. However, you should realise that that last few bits in any Analogue to Digital Converter (ADC) is where the noise lives. As a general rule (in electronics) this noise is overcome by a) averaging over several samples (not usually possible with a camera) or throwing away the last few bits - which is what the jpeg conversion process does anyway.

It's not too suprising that the A1 and A2 differ - the sensors are from different manufacturers. It's possible (even probable?) that the A1 has slightly better colour depth - but unless you very carefully compare the RAW images you won't be able to tell whether those extra 2bits are just random noise - or useful data.

And this is why (as Eliezer pointed out) you need to aim to get the basic exposure right in the first place. You can shift your conversion down into the 12bit range a bit - but the further you push it, the more you will approach noise. Personally I find that shifting more the 1eV gets difficult, although occaisionally you can go the full 2eV.

I would recommend you get Real World Camera RAW with Adobe Photoshop CS by Bruce Fraser as it sounds like you're using PS CS ACR. I've only read a few chapters so far, but they have greatly changed my understanding and use of ACR.

ovatsus said:

Humm, so changing the exposure compensation in the RAW converter is
the same as adjusting the levels after the conversion, if we use a
lossless adn 16bit final format, right?

Click to expand...

Yes and no. Exposure is a poor term for what actually happens in ACR. You are right in that ACR Exposure sets the white point just like in levels. Similarly, ACR Shadows sets the black point just like in levels. However, it may be much better to do those adjustments in ACR, depending on how much you want to change things of course. Apparently ACR uses a linear colour space before you convert to SRGB or whatever, so ACR does a better job of using available colour depth to adjust the scene than using levels or curves after conversion. Bruce recommends using ACR for contrast and brightness as well if you adjust exposure by more than 0.75Ev.

Another advantage of ACR is its ability to recover blown highlights if only one or maybe two channels are blown. All of this stuff is explained in the book, along with heaps of other very useful things. Bruce recommends slightly overexposing so that you get a bit more shadow information while being able to recover blown details as long as they aren't too blown.

--
Cheers from down under in Adelaide, South Australia, Australia



My PBase galleries since 16 Apr 04: http://www.pbase.com/john_down_under
Konica-Minolta Challenges, Faces and FAx: http://www.pbase.com/mtf_foto_studies

While the manual says that the A1 has 14bit RAW images, there is some doubt about whether that is true and, unless you know for sure, you would have to assume that it may be a mistake. In any case, 12 bits is pretty good when compared with the 8 bits you get from shooting JPEGs and provides plenty of scope to do serious manipulation during PP. 12 bits seems to be standard for RAW in most cameras AFAIK.

--
Cheers from down under in Adelaide, South Australia, Australia



My PBase galleries since 16 Apr 04: http://www.pbase.com/john_down_under
Konica-Minolta Challenges, Faces and FAx: http://www.pbase.com/mtf_foto_studies

Thanks for the book recomendation

One thing I still wonder. Does the sRGB and AdobeRGB color spaces apply to 16-bit images? If both are different partitions of the colour into 8-bits...

John down under said:

ovatsus said:

Humm, so changing the exposure compensation in the RAW converter is
the same as adjusting the levels after the conversion, if we use a
lossless adn 16bit final format, right?

Click to expand...

Yes and no. Exposure is a poor term for what actually happens in
ACR. You are right in that ACR Exposure sets the white point just
like in levels. Similarly, ACR Shadows sets the black point just
like in levels. However, it may be much better to do those
adjustments in ACR, depending on how much you want to change things
of course. Apparently ACR uses a linear colour space before you
convert to SRGB or whatever, so ACR does a better job of using
available colour depth to adjust the scene than using levels or
curves after conversion. Bruce recommends using ACR for contrast
and brightness as well if you adjust exposure by more than 0.75Ev.

Another advantage of ACR is its ability to recover blown highlights
if only one or maybe two channels are blown. All of this stuff is
explained in the book, along with heaps of other very useful
things. Bruce recommends slightly overexposing so that you get a
bit more shadow information while being able to recover blown
details as long as they aren't too blown.

--
Cheers from down under in Adelaide, South Australia, Australia



My PBase galleries since 16 Apr 04:
http://www.pbase.com/john_down_under
Konica-Minolta Challenges, Faces and FAx:
http://www.pbase.com/mtf_foto_studies

Click to expand...

John down under said:

While the manual says that the A1 has 14bit RAW images, there is
some doubt about whether that is true and, unless you know for
sure, you would have to assume that it may be a mistake. In any
case, 12 bits is pretty good when compared with the 8 bits you get
from shooting JPEGs and provides plenty of scope to do serious
manipulation during PP. 12 bits seems to be standard for RAW in
most cameras AFAIK.

Click to expand...

Minolta put out a file called DIMAGEA1Backgrounder.pdf, in which it states:

"...The DiMAGE A1 uses 14-bit A/D conversion to record the richness and infinite subtleties of light. 14-bit A/D conversion can distinguish 16,384 levels in each color channel. This is a fourfold increase over the fidelity of 12-bit systems. Images simply come alive..."

I take this to mean that the A/D convertor is 14-bits, but it seems to me that the actual RAW format is then converted to 12-bits, as for the other cameras in the series. Whether this latter conversion is done in such a way as to preserve any of the 14-bit advantage (e.g. wioth dither), or whether the last two bits are just igniored (truncated), I do not know. I also do not honestly know whether these 2 bits contain any really useful information -- that depends upon the CCD.

It may well be the case, however, that the 12th bit linearity of a 14-bit convertor was better than that of available 12-bit convertors at the time the A1 was designed, and this might be the basis for Minolta's claim.

david

ovatsus said:

Thanks for the book recomendation

One thing I still wonder. Does the sRGB and AdobeRGB color spaces
apply to 16-bit images? If both are different partitions of the
colour into 8-bits...

Click to expand...

I'm not sure I understand what you're asking. In fact, I'm not sure you know what you're asking! LOL Another book I got very recently is Real World Color Management Second Edition by Bruce Fraser, etc. I have a basic understanding of colour space, but I hope to have a much better understanding once I've read the book.

One answer to your question is that sRGB and AdobeRGB can be 8 or 16 bit images. You can choose to convert MRW to 16 or 8 bit sRGB, 16 or 8 bit Adobe RGB or other colour spaces/depths if you want. AFAIK, you will only get 12 bits of real information (or is it 14 for the A1?? ;o)) when you convert MRW to a 16 bit image, but that's a lot more than 8. You can then process as a 16 bit image and save for displaying on the web as an 8 bit image. So far, I'm happy to work with sRGB rather than AdobeRGB and that gives pretty good results with very little hassle to post on the web.

--
Cheers from down under in Adelaide, South Australia, Australia



My PBase galleries since 16 Apr 04: http://www.pbase.com/john_down_under
Konica-Minolta Challenges, Faces and FAx: http://www.pbase.com/mtf_foto_studies

ovatsus said:

2) And while on exposure compensation, what does the flash exposure
compensation really does? Modify the intensity of the flash, act as
an independent exposure compensation setting that is used when the
flash is used, or a combination of both? I'm inclined to think it
is the first, as in flash manual mode the setting is gone, but i
want to make sure it'ss not the third.

Click to expand...

It modifies ONLY the flash output (ie: intensity). This is useful for when you want a very mild amount of fill flash when shooting a strongly backlit photo.

Larry

John down under said:

ovatsus said:

Thanks for the book recomendation

One thing I still wonder. Does the sRGB and AdobeRGB color spaces
apply to 16-bit images? If both are different partitions of the
colour into 8-bits...

Click to expand...

I'm not sure I understand what you're asking. In fact, I'm not sure
you know what you're asking! LOL Another book I got very recently
is Real World Color Management Second Edition by Bruce Fraser, etc.
I have a basic understanding of colour space, but I hope to have a
much better understanding once I've read the book.

One answer to your question is that sRGB and AdobeRGB can be 8 or
16 bit images. You can choose to convert MRW to 16 or 8 bit sRGB,
16 or 8 bit Adobe RGB or other colour spaces/depths if you want.
AFAIK, you will only get 12 bits of real information (or is it 14
for the A1?? ;o)) when you convert MRW to a 16 bit image, but
that's a lot more than 8. You can then process as a 16 bit image
and save for displaying on the web as an 8 bit image. So far, I'm
happy to work with sRGB rather than AdobeRGB and that gives pretty
good results with very little hassle to post on the web.

Click to expand...

Yes, your right, I guess I don't really know what I'm asking I didn't really understand what colour space is.

My though was: 8bits can't cover the full amount of colors fisically possible, so, we have to choose wich to represent. And sRGB chooses one and AdobeRGB other, the second having a wider range, but less colors in a interval.
So it seems it's not exactly like that, have to investigate further...

Thanks

ovatsus said:

John down under said:

ovatsus said:

Thanks for the book recomendation

One thing I still wonder. Does the sRGB and AdobeRGB color spaces
apply to 16-bit images? If both are different partitions of the
colour into 8-bits...

Click to expand...

I'm not sure I understand what you're asking. In fact, I'm not sure
you know what you're asking! LOL Another book I got very recently
is Real World Color Management Second Edition by Bruce Fraser, etc.
I have a basic understanding of colour space, but I hope to have a
much better understanding once I've read the book.

One answer to your question is that sRGB and AdobeRGB can be 8 or
16 bit images. You can choose to convert MRW to 16 or 8 bit sRGB,
16 or 8 bit Adobe RGB or other colour spaces/depths if you want.
AFAIK, you will only get 12 bits of real information (or is it 14
for the A1?? ;o)) when you convert MRW to a 16 bit image, but
that's a lot more than 8. You can then process as a 16 bit image
and save for displaying on the web as an 8 bit image. So far, I'm
happy to work with sRGB rather than AdobeRGB and that gives pretty
good results with very little hassle to post on the web.

Click to expand...

Yes, your right, I guess I don't really know what I'm asking I
didn't really understand what colour space is.

My though was: 8bits can't cover the full amount of colors
fisically possible, so, we have to choose wich to represent. And
sRGB chooses one and AdobeRGB other, the second having a wider
range, but less colors in a interval.
So it seems it's not exactly like that, have to investigate further...

Thanks

Click to expand...

The optimal flow in edit photos
1) Make what you can in the raw-editor (psCS camera raw) in 16bit
2) fine-tune and filter in psCS in 16bit
3) Make the leftovers in plug-ins or other programs, best in 16 bits
4) make it to 8bits for saving as jpg
Use adobeRGB to use for prints - from start
Use sRGB to use for screen - from start
http://jens.noehr.person.emu.dk/
--
Jens Noer

David-P said:

John down under said:

While the manual says that the A1 has 14bit RAW images, there is
some doubt about whether that is true and, unless you know for
sure, you would have to assume that it may be a mistake. In any
case, 12 bits is pretty good when compared with the 8 bits you get
from shooting JPEGs and provides plenty of scope to do serious
manipulation during PP. 12 bits seems to be standard for RAW in
most cameras AFAIK.

Click to expand...

Minolta put out a file called DIMAGEA1Backgrounder.pdf, in which it
states:

"...The DiMAGE A1 uses 14-bit A/D conversion to record the richness
and infinite subtleties of light. 14-bit A/D conversion can
distinguish 16,384 levels in each color channel. This is a fourfold
increase over the fidelity of 12-bit systems. Images simply come
alive..."

I take this to mean that the A/D convertor is 14-bits, but it seems
to me that the actual RAW format is then converted to 12-bits, as
for the other cameras in the series. Whether this latter
conversion is done in such a way as to preserve any of the 14-bit
advantage (e.g. wioth dither), or whether the last two bits are
just igniored (truncated), I do not know. I also do not honestly
know whether these 2 bits contain any really useful information --
that depends upon the CCD.

It may well be the case, however, that the 12th bit linearity of a
14-bit convertor was better than that of available 12-bit
convertors at the time the A1 was designed, and this might be the
basis for Minolta's claim.

david

Click to expand...

--
Cheers from down under in Adelaide, South Australia, Australia



My PBase galleries since 16 Apr 04: http://www.pbase.com/john_down_under
Konica-Minolta Challenges, Faces and FAx: http://www.pbase.com/mtf_foto_studies

I'll have to update the colour space section in our FAx to make it more informative and useful. In the meantime, here's some basic info.

The pixel values (numbers) in an image must represent colours in the real world. The colour space defines how the pixel values map to real colours. sRGB maps to a narrower colour space or range of colours than AdobeRGB does, although not hugely different. If you're using the same bit depth to represent the colours in both colour spaces (or in any colour space), you can only have the same number of different colours. For that reason, the colours in AdobeRGB are slightly further apart (slightly more different from each other) than are sRGB colours. Remember, the same number of different colours fills the colour space in ech case, so the smaller colour space means the colour values are crammed in more tightly.

sRGB may be prefereable for displaying on computer monitors as almost none of them can display the full range of AdobeRGB colours, but they can all display the full range of sRGB colours.

Another application, even for print, where sRGB may be better is for portraits where you don't necessarily need a wide range of colours and in fact the finer gradient resulting from adjacent colours being more similar to each other than for AdobeRGB might produce a better result than you can get from AdobeRGB. However, I doubt there's a lot in it.

The bit depth is an independent variable.The size/boundary of the colour space is the same regardless of the bit depth being used for an image in that colour space (eg 8 bits/channel, 12/14 bits/channel or 16 bits/channel). However, an 8 bit image in a given colour space is obviously going to have adjacent colours much further separated (much more different from each other) than say a 12 bit image in the same colour space. Adjacent colour values in an 8 bit image will represent two different colours. The same two colours for a 12 bit image in that same colour space are just that, exactly the same colours as for the 8 bit image. However, they are not adjacent colours. In fact, the colours are 16 different colours apart, ie 2^(12-8) = 2^4 = 16. All those extra colours can make for a much smoother and more accurate colour gradient, ie a smoother and more realistic image.

ovatsus said:

Yes, your right, I guess I don't really know what I'm asking I
didn't really understand what colour space is.

My though was: 8bits can't cover the full amount of colors
fisically possible, so, we have to choose wich to represent. And
sRGB chooses one and AdobeRGB other, the second having a wider
range, but less colors in a interval.
So it seems it's not exactly like that, have to investigate further...

Thanks

Click to expand...

--
Cheers from down under in Adelaide, South Australia, Australia



My PBase galleries since 16 Apr 04: http://www.pbase.com/john_down_under
Konica-Minolta Challenges, Faces and FAx: http://www.pbase.com/mtf_foto_studies