In article , Dave Martindale
wrote:

(Toby Thain) writes:

I've tried. Their engineer insists that it's 30x faster to work with
15 bit quantities than 16 bit ones.

When your source data was probably from a 12-bit ADC, or maybe 14-bit,
working with 15 significant bits may indeed be completely adequate. And
there *are* advantages to using a representation that has some headroom
for "whiter than white" without overflow, and where the representation
for "1.0" is a power of 2.

But the couple of most recent comments in this thread are about the fact
that Photoshop's greyscale doesn't even seem to have 15 significant
bits, unlike the RGB representation.

The color mode doesn't matter - it's still 16 bit data (0..32768).

Chris

In message ,
Chris Cox wrote:

Without your original data to test, I can't even guess what went wrong.

You don't need my original data.

Any image in "16 bit greyscale" mode has all kinds of numbers between 0
and 32768 missing, and not possible no matter hown much you blur or
interpolate. "16 bit greyscale" is about 13.5 bit greyscale.
--


John P Sheehy

Chris Cox wrote:

In article , Dave Martindale
wrote:


(Toby Thain) writes:


I've tried. Their engineer insists that it's 30x faster to work with
15 bit quantities than 16 bit ones.

When your source data was probably from a 12-bit ADC, or maybe 14-bit,
working with 15 significant bits may indeed be completely adequate. And
there *are* advantages to using a representation that has some headroom
for "whiter than white" without overflow, and where the representation
for "1.0" is a power of 2.

But the couple of most recent comments in this thread are about the fact
that Photoshop's greyscale doesn't even seem to have 15 significant
bits, unlike the RGB representation.


The color mode doesn't matter - it's still 16 bit data (0..32768).

Hi Chris...

0..32767 or 1..32768

You just can't have it both ways :)

Ken



Chris

Chris Cox wrote:

In article , Dave Martindale
wrote:


(Toby Thain) writes:


I've tried. Their engineer insists that it's 30x faster to work with
15 bit quantities than 16 bit ones.

When your source data was probably from a 12-bit ADC, or maybe 14-bit,
working with 15 significant bits may indeed be completely adequate. And
there *are* advantages to using a representation that has some headroom
for "whiter than white" without overflow, and where the representation
for "1.0" is a power of 2.

But the couple of most recent comments in this thread are about the fact
that Photoshop's greyscale doesn't even seem to have 15 significant
bits, unlike the RGB representation.


The color mode doesn't matter - it's still 16 bit data (0..32768).

Hi Chris...

0..32767 or 1..32768

You just can't have it both ways :)

Ken



Chris

Chris Cox writes:

I've tried. Their engineer insists that it's 30x faster to work with
15 bit quantities than 16 bit ones.

Which is correct (for 0..32768 representation versus 0..65535
representation).

Perhaps this is offtopic, and perhaps you can't answer it without
revealing proprietary information, but can you explain why 15-bit
computation should be so much faster than 16-bit? (If there's a
publication somewhere you could point me to, that would be great.)
I've thought about this for a few minutes, I haven't been able to
think of an obvious reason, and now I'm curious.

Feel free to email me if you think this wouldn't be interesting to
anyone else.

Chris Cox writes:

I've tried. Their engineer insists that it's 30x faster to work with
15 bit quantities than 16 bit ones.

Which is correct (for 0..32768 representation versus 0..65535
representation).

Perhaps this is offtopic, and perhaps you can't answer it without
revealing proprietary information, but can you explain why 15-bit
computation should be so much faster than 16-bit? (If there's a
publication somewhere you could point me to, that would be great.)
I've thought about this for a few minutes, I haven't been able to
think of an obvious reason, and now I'm curious.

Feel free to email me if you think this wouldn't be interesting to
anyone else.

Matt Austern wrote:

Chris Cox writes:


I've tried. Their engineer insists that it's 30x faster to work with
15 bit quantities than 16 bit ones.

Which is correct (for 0..32768 representation versus 0..65535
representation).


Perhaps this is offtopic, and perhaps you can't answer it without
revealing proprietary information, but can you explain why 15-bit
computation should be so much faster than 16-bit? (If there's a
publication somewhere you could point me to, that would be great.)
I've thought about this for a few minutes, I haven't been able to
think of an obvious reason, and now I'm curious.

Feel free to email me if you think this wouldn't be interesting to
anyone else.


Hi Matt...

Nor can I see even the slightest difference. None at all.

So - I suspect that we're looking at it from the wrong
end. Suspect it's the a/d converter that could be the
bottleneck?

8 bits are common; 15 bit's are common. 18 bit
are available but seldom used. Never heard of 16.
Maybe that's it?

Ken

Matt Austern wrote:

Chris Cox writes:


I've tried. Their engineer insists that it's 30x faster to work with
15 bit quantities than 16 bit ones.

Which is correct (for 0..32768 representation versus 0..65535
representation).


Perhaps this is offtopic, and perhaps you can't answer it without
revealing proprietary information, but can you explain why 15-bit
computation should be so much faster than 16-bit? (If there's a
publication somewhere you could point me to, that would be great.)
I've thought about this for a few minutes, I haven't been able to
think of an obvious reason, and now I'm curious.

Feel free to email me if you think this wouldn't be interesting to
anyone else.


Hi Matt...

Nor can I see even the slightest difference. None at all.

So - I suspect that we're looking at it from the wrong
end. Suspect it's the a/d converter that could be the
bottleneck?

8 bits are common; 15 bit's are common. 18 bit
are available but seldom used. Never heard of 16.
Maybe that's it?

Ken

Ken Weitzel writes:

Matt Austern wrote:

Chris Cox writes:

I've tried. Their engineer insists that it's 30x faster to work with
15 bit quantities than 16 bit ones.

Which is correct (for 0..32768 representation versus 0..65535
representation).
Perhaps this is offtopic, and perhaps you can't answer it without
revealing proprietary information, but can you explain why 15-bit
computation should be so much faster than 16-bit? (If there's a
publication somewhere you could point me to, that would be great.)
I've thought about this for a few minutes, I haven't been able to
think of an obvious reason, and now I'm curious.
Feel free to email me if you think this wouldn't be interesting to
anyone else.


Hi Matt...

Nor can I see even the slightest difference. None at all.

So - I suspect that we're looking at it from the wrong
end. Suspect it's the a/d converter that could be the
bottleneck?

Nope. If Chris says 16-bit image processing in Photoshop would be much
slower than 15, I have no doubt that he's right. I just don't know
why. I can easily believe there's some subtle algorithmic issue that
I haven't thought of. For that matter, I can easily believe there's
some glaringly obvious algorithmic issue I haven't thought of. I'm
just curious what it might be.

Ken Weitzel writes:

Matt Austern wrote:

Chris Cox writes:

I've tried. Their engineer insists that it's 30x faster to work with
15 bit quantities than 16 bit ones.

Which is correct (for 0..32768 representation versus 0..65535
representation).
Perhaps this is offtopic, and perhaps you can't answer it without
revealing proprietary information, but can you explain why 15-bit
computation should be so much faster than 16-bit? (If there's a
publication somewhere you could point me to, that would be great.)
I've thought about this for a few minutes, I haven't been able to
think of an obvious reason, and now I'm curious.
Feel free to email me if you think this wouldn't be interesting to
anyone else.


Hi Matt...

Nor can I see even the slightest difference. None at all.

So - I suspect that we're looking at it from the wrong
end. Suspect it's the a/d converter that could be the
bottleneck?

Nope. If Chris says 16-bit image processing in Photoshop would be much
slower than 15, I have no doubt that he's right. I just don't know
why. I can easily believe there's some subtle algorithmic issue that
I haven't thought of. For that matter, I can easily believe there's
some glaringly obvious algorithmic issue I haven't thought of. I'm
just curious what it might be.