Velocity Reviews - Computer Hardware Reviews

Velocity Reviews > Newsgroups > Computing > Digital Photography > "digital" flash mode (no actual flash fired) HP945

Reply
Thread Tools

"digital" flash mode (no actual flash fired) HP945

 
 
David Bindle
Guest
Posts: n/a
 
      07-12-2004
I was just reading what Bob Atkins had to say about the "digital" flash mode
on the HP945.
He demonstrates with a mountain scene. It doesn't actually fire the flash at
all, but it lightens up darker parts of the scene. For some purposes, it
looks like a good idea to me.

I wonder, could this technique effectively replace the split neutral density
filters that are commonly used in mountain scenics where the valleys are so
much darker than the white capped mountains? (Of course, only in digital
cameras...)

Can you achieve this with basically all digital cameras by just reducing the
contrast settings?

Are there any other cameras besides the HP945 that has a similar flash, or
program setting?

I wonder how well this would work for flashless portraits near full
telephoto (~300mm equiv.) to bring out shadow detail in the eyes on a sunny
day.

Thanks for any help or info you can provide.

David


 
Reply With Quote
 
 
 
 
Sabineellen
Guest
Posts: n/a
 
      07-13-2004
>
>I was just reading what Bob Atkins had to say about the "digital" flash mode
>on the HP945.
>He demonstrates with a mountain scene. It doesn't actually fire the flash at
>all, but it lightens up darker parts of the scene. For some purposes, it
>looks like a good idea to me.
>
>I wonder, could this technique effectively replace the split neutral density
>filters that are commonly used in mountain scenics where the valleys are so
>much darker than the white capped mountains? (Of course, only in digital
>cameras...)
>
>Can you achieve this with basically all digital cameras by just reducing the
>contrast settings?
>
>Are there any other cameras besides the HP945 that has a similar flash, or
>program setting?
>
>I wonder how well this would work for flashless portraits near full
>telephoto (~300mm equiv.) to bring out shadow detail in the eyes on a sunny
>day.
>
>Thanks for any help or info you can provide.
>
>David
>
>


Hi... I had the HP 945 for a week now and it's an easily likeable camera,
though i have not used the digital flash function and don't really have
intention to anytime soon.

From what i read, and I assume you probably read it itoo, the advantage of the
one done on the HP is that the camera does it directly on the fresh CCD data,
which means it'll probably do it better than JPEG editing on the computer. I
think it's different from contrast settings in cameras, because this one seems
a little "intelligent", if that's the right word. Basically, reviewers seemed
impressed with it, which gave me the impression they probably didn't expect it
to work, but it did.

I have no idea how it'd perform on a sunny day portrait, but i have the vague
impression that i read somewhere that it can be used for such situations of
harsh shadows on sunny days and that it does actually work. That said though,
the camera has up to +/-3 in exposure compensation which is the usual trick,
plus a spot metering mode. Those two are what is usually used for such
situations, at least in 35mm photography.

I also vaguely remember reading one user review that said indoors it could turn
a black hair gray, but that was the only negative comment i read regarding it.

Maybe i ought to try it for you.

The only other camera that has it is the newer HP camera, HP Photosmart R707,
just out lately.




 
Reply With Quote
 
 
 
 
Michael Meissner
Guest
Posts: n/a
 
      07-13-2004
"David Bindle" <(E-Mail Removed)> writes:

> I was just reading what Bob Atkins had to say about the "digital" flash mode
> on the HP945.
> He demonstrates with a mountain scene. It doesn't actually fire the flash at
> all, but it lightens up darker parts of the scene. For some purposes, it
> looks like a good idea to me.
>
> I wonder, could this technique effectively replace the split neutral density
> filters that are commonly used in mountain scenics where the valleys are so
> much darker than the white capped mountains? (Of course, only in digital
> cameras...)


The problem is most digital cameras have a lot less range than print films do,
between dark and light. Slide films have the same sort of problems, and the
solution is typically the same (expose to avoid burning out the highlights and
bring out shadow detail in post processing). In the most common case of JPEG,
there are only 8 bits (values 0-255) for each color per pixel. With some
cameras by going to the RAW formats, you get a bit more range. I suspect that
in general you will get better results under a photo editor like photoshop or
gimp than with the camera doing the processing.

--
Michael Meissner
email: http://www.velocityreviews.com/forums/(E-Mail Removed)
http://www.the-meissners.org
 
Reply With Quote
 
Paul H.
Guest
Posts: n/a
 
      07-13-2004

"David Bindle" <(E-Mail Removed)> wrote in message
news:ccv0c6$fm4$(E-Mail Removed)...
> I was just reading what Bob Atkins had to say about the "digital" flash

mode
> on the HP945.
> He demonstrates with a mountain scene. It doesn't actually fire the flash

at
> all, but it lightens up darker parts of the scene. For some purposes, it
> looks like a good idea to me.
>
> I wonder, could this technique effectively replace the split neutral

density
> filters that are commonly used in mountain scenics where the valleys are

so
> much darker than the white capped mountains? (Of course, only in digital
> cameras...)
>
> Can you achieve this with basically all digital cameras by just reducing

the
> contrast settings?
>
> Are there any other cameras besides the HP945 that has a similar flash, or
> program setting?
>
> I wonder how well this would work for flashless portraits near full
> telephoto (~300mm equiv.) to bring out shadow detail in the eyes on a

sunny
> day.
>
> Thanks for any help or info you can provide.
>
> David




David, while it's neat that the HP-945 provides this function in-camera, the
same sort of "digital flash" can be applied out of the camera using any
fairly sophisticated photo editor such as Photoshop or Photoshop Elements.
The technique is called "constrast masking" and is used to selectively
brighten darker portions of an image while leaving the brighter parts
relatively untouched. Actually, "contrast masking" is a bit of a misnomer;
"luminance enhancement" would probably be a better term.

Here's how it's done in Photoshop Elements. With an image loaded into the
program,

1) Duplicate the image layer and make the duplicate layer the active layer
2) De-saturate the layer (ctrl-shift-U)
3) Invert the layer (ctrl-I)
4) Set the blending mode of the layer to "Overlay" or "Soft Light". Overlay
provides greater contrast.
5) Set the layer transparency to around 80%.
6) Apply a Gaussian blur to the layer. Start with something around 50 and
work up/down from there. The purpose of this blurring step is to reduce the
severity of the transitions between light and dark areas in the photo to
avoid that awful cut-and-paste look.
7) Optional--Play around with steps 4, 5 and 6 to get the most pleasing
result.
Flatten the image (ctrl-E)
9) Optional--Adjust Levels or Brightness/Contrast, if desired
10) Save the picture under a different name.

If you've never done this before, it might sound complicated, but it really
isn't in practice. Just a few keystrokes/button-presses and you've got an
amazingly enhanced picture in many cases and you can often even turn
throw-away shots into real keepers using this technique. ("Yes--now you can
turn TRASH into CASH!" as an infomercial might put it.) Also, if you're
using full Photoshop, not Elements, you can turn this process into an
action. Finally, to answer your other question, enhancing
brightness/contrast on a digital camera simply can't duplicate the
digital-flash/contrast-masking effect.

I'd rather do my "digital flashing" outside the camera anyway since I have
far more control over the finished result, though as I said above, it *is*
an interesting feature to include in a digital camera. BTW,
www.steves-digicams.com has an HP-945 review with a couple of digital-flash
examples along with their normally-shot counterparts so you can try out
contrast masking on your own to see how it compares to HP's implementation.

Hope this helps.








 
Reply With Quote
 
Charles Jones
Guest
Posts: n/a
 
      07-14-2004
In article <(E-Mail Removed)>,
(E-Mail Removed)m says...

> Actually, "contrast masking" is a bit of a misnomer;
> "luminance enhancement" would probably be a better term.
>
> Here's how it's done in Photoshop Elements. With an image loaded into the
> program,


[snip excellent step-by-step]

> Hope this helps.


Indeed! Thanks for posting this; it's a great tool to add to my quiver.

--
Charles Jones -- Loveland, Colorado
ICQ: 29610755
AIM: LovelandCharles
Y!M: charlesjonesathpcom
MSN: (E-Mail Removed)
 
Reply With Quote
 
andrew29@littlepinkcloud.invalid
Guest
Posts: n/a
 
      07-24-2004
Michael Meissner <(E-Mail Removed)> wrote:

> The problem is most digital cameras have a lot less range than print
> films do, between dark and light. Slide films have the same sort of
> problems, and the solution is typically the same (expose to avoid
> burning out the highlights and bring out shadow detail in post
> processing). In the most common case of JPEG, there are only 8 bits
> (values 0-255) for each color per pixel.


Yeah, but usually those 8 bits represent a value that you raise to the
power 2.2 to get luminance. When you work that out, 8 bits is a heck
of a dynamic range.

From what I can see, digital sensors (or at least those in DSLRs) have
more dynamic range than transparencies but less than print film. It
may be that with a drum scanner you can squeeze a bit more out of a
transparency.

> With some cameras by going to the RAW formats, you get a bit more
> range. I suspect that in general you will get better results under
> a photo editor like photoshop or gimp than with the camera doing the
> processing.


Yup.

Andrew.
 
Reply With Quote
 
Michael Meissner
Guest
Posts: n/a
 
      07-24-2004
(E-Mail Removed)lid writes:

> Michael Meissner <(E-Mail Removed)> wrote:
>
> > The problem is most digital cameras have a lot less range than print
> > films do, between dark and light. Slide films have the same sort of
> > problems, and the solution is typically the same (expose to avoid
> > burning out the highlights and bring out shadow detail in post
> > processing). In the most common case of JPEG, there are only 8 bits
> > (values 0-255) for each color per pixel.

>
> Yeah, but usually those 8 bits represent a value that you raise to the
> power 2.2 to get luminance. When you work that out, 8 bits is a heck
> of a dynamic range.


The classic counterexample is shooting weddings, and trying to get detail in
both the bride's white dress and the groom's black tux. From what I've read,
most digital cameras give you the range of slide film, but print film still
gives more latitude.

> From what I can see, digital sensors (or at least those in DSLRs) have
> more dynamic range than transparencies but less than print film. It
> may be that with a drum scanner you can squeeze a bit more out of a
> transparency.


I believe in general you only get that extended range in DSLRs when you use raw
mode, which for many cameras can give you 10-12 bits of precision. I believe
some of the newer Fujis are trying to address the dynamic range problem by
having two sensors for the high and low values. And even the Sigmas are trying
to attack the problem, its a pity that the implementation leaves a lot to be
desired compared to the theory.

I suspect within 5 years or so, there will be a shift to using 16-bit formats
in cameras. As bragging rights, I would hope the megapixel race is nearly
over, and manufacturers start concentrating on higher ISO values (with less
noise) and more tonality.

--
Michael Meissner
email: (E-Mail Removed)
http://www.the-meissners.org
 
Reply With Quote
 
andrew29@littlepinkcloud.invalid
Guest
Posts: n/a
 
      07-25-2004
Michael Meissner <(E-Mail Removed)> wrote:
> (E-Mail Removed)lid writes:


>> Michael Meissner <(E-Mail Removed)> wrote:
>>
>> > The problem is most digital cameras have a lot less range than print
>> > films do, between dark and light. Slide films have the same sort of
>> > problems, and the solution is typically the same (expose to avoid
>> > burning out the highlights and bring out shadow detail in post
>> > processing). In the most common case of JPEG, there are only 8 bits
>> > (values 0-255) for each color per pixel.

>>
>> Yeah, but usually those 8 bits represent a value that you raise to the
>> power 2.2 to get luminance. When you work that out, 8 bits is a heck
>> of a dynamic range.


> The classic counterexample is shooting weddings, and trying to get detail in
> both the bride's white dress and the groom's black tux.


But that's not really a counterexample: there's no way that the ratio
of tux/dress exceeds the dynamic range of a JPEG encoded at gamma 2.2.
The limiting factor is still sensor noise.

> From what I've read, most digital cameras give you the range of
> slide film, but print film still gives more latitude.


>> From what I can see, digital sensors (or at least those in DSLRs)
>> have more dynamic range than transparencies but less than print
>> film. It may be that with a drum scanner you can squeeze a bit
>> more out of a transparency.


> I believe in general you only get that extended range in DSLRs when
> you use raw mode, which for many cameras can give you 10-12 bits of
> precision.


What raw gives you is linear encoding, at least with CCDs. That gives
you an opportunity to use curves somehow to squeeze that huge dynamic
range into that of a print -- which has Dmax 1.2 if you're lucky.

> I suspect within 5 years or so, there will be a shift to using
> 16-bit formats in cameras.


Actively cooled sensors, maybe?

Andrew.
 
Reply With Quote
 
Michael Meissner
Guest
Posts: n/a
 
      07-28-2004
(E-Mail Removed)lid writes:

> Michael Meissner <(E-Mail Removed)> wrote:
> > (E-Mail Removed)lid writes:

>
> >> Michael Meissner <(E-Mail Removed)> wrote:
> >>
> >> > The problem is most digital cameras have a lot less range than print
> >> > films do, between dark and light. Slide films have the same sort of
> >> > problems, and the solution is typically the same (expose to avoid
> >> > burning out the highlights and bring out shadow detail in post
> >> > processing). In the most common case of JPEG, there are only 8 bits
> >> > (values 0-255) for each color per pixel.
> >>
> >> Yeah, but usually those 8 bits represent a value that you raise to the
> >> power 2.2 to get luminance. When you work that out, 8 bits is a heck
> >> of a dynamic range.

>
> > The classic counterexample is shooting weddings, and trying to get detail in
> > both the bride's white dress and the groom's black tux.

>
> But that's not really a counterexample: there's no way that the ratio
> of tux/dress exceeds the dynamic range of a JPEG encoded at gamma 2.2.
> The limiting factor is still sensor noise.


You sound like you know more about it than I do, but I do know weddings can
stress most cameras, and usually you have to underexpose digitals so that the
highlights aren't blown.

> > From what I've read, most digital cameras give you the range of
> > slide film, but print film still gives more latitude.

>
> >> From what I can see, digital sensors (or at least those in DSLRs)
> >> have more dynamic range than transparencies but less than print
> >> film. It may be that with a drum scanner you can squeeze a bit
> >> more out of a transparency.

>
> > I believe in general you only get that extended range in DSLRs when
> > you use raw mode, which for many cameras can give you 10-12 bits of
> > precision.

>
> What raw gives you is linear encoding, at least with CCDs. That gives
> you an opportunity to use curves somehow to squeeze that huge dynamic
> range into that of a print -- which has Dmax 1.2 if you're lucky.
>
> > I suspect within 5 years or so, there will be a shift to using
> > 16-bit formats in cameras.


I was just speculating that if you go past 8 bits of precision, the next format
is 16 bits -- even if you only have 12 bits worth of precision from the camera.

> Actively cooled sensors, maybe?


Perhaps.

--
Michael Meissner
email: (E-Mail Removed)
http://www.the-meissners.org
 
Reply With Quote
 
 
 
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
mmm-mode, python-mode and doctest-mode? John J Lee Python 3 12-01-2005 08:35 PM
SONY VS HP945 the unspeakable truth! youlookhot@gmail.com Digital Photography 7 12-01-2005 07:15 AM
HP945 Photosmart Camera. Gordon Digital Photography 7 07-08-2005 07:57 PM
Safe Mode (?) - It is meant to be normal mode but looks like safe mode English Patient Computer Support 3 10-03-2004 11:10 PM
HP945? Jerry Ennis Digital Photography 4 07-09-2004 09:37 PM



Advertisments