DSLR sales static, mirrorless heavy growth?

Trevor wrote:
"Alan Browne" wrote in message

That's why most of us save in raw. But not all people do and so they lose
information. It's unavoidable.

Exactly my point all along.

You lose information any time you save in JPEG (e.g. after
converting your RAWs). It's called lossy compression for a reason.

Whether that information is useful to you is a different question.

On the other hand there are a lot of photographers who are satisfied with
the in-camera result or at least with the narrow range of change they can
do with the in-camera JPG. There's nothing 'wrong' with that if it meets
their needs.

Sure, but claiming they don't lose dynamic range (or only 1 or 2 bits) is
what I objected to.

It's trivial to devise a linear 5 bit format that has more dynamic
range than a 14 bit RAW from a camera. (Just use bigger steps!).

So what's the reason an 8 bit format with variable step size
can't encode as much DR as your camera's RAW? You claim it
cannot. So ... prove it.

-Wolfgang

On 2012-07-23 00:36 , Trevor wrote:
"Alan Browne" wrote in message
...
A good JPG doesn't "lose" dynamic range because of the number of bits,
it
is a log compressed format and as such can potentially represent the
full
DR of an image - it does lose fine dynamic resolution, of course.

I'm unaware of any camera that can adjust curves before saving to jpeg,
and

They have no choice but to do so.
1. There is no way for the camera to go from its internal represented
image (aka raw) to the displayed image on the camera display or to the
saved file without applying curves.

What is displayed on the little LCD screen is not as important as what is
saved to file however.


2. You don't select the curves in-camera, it's pre-programmed.

Exactly, which *IS* the problem!

No it isn't.

.. For the satisfied-with-in-camera-JPG-user it is a solution.

.. For the raw user it is irrelevant.

My camera has several modes of JPEG "saving" which of course changes the
curve (and other parameters of the save) for each flavour. (I ignore these
generally as I do everything from the raw).

Right, me too.


3. Note that many cameras have camera-direct-to-print capability and so
the use of curves is not only unavoidable but absolutely necessary to get
a useful image.

Sure, if that's what you are after. I have no use for that either.
Many people are happy with photo's from their iphones, and RAW files sure
aren't necessary (or posible) with those. Neither could you claim they had
"12-14 bits" dynamic range :-)

That's another discussion. There are a few cameras with such range
(Hasselblad digital cameras are generally 16 bit). Even after throwing
away 2 bits for noise you have 14 bits.

I sure as hell find it easier to do it at my leasure afterwards, than try
to
do it with each shot as I take it, even if they did, so I don't have any
use
for saving 8 bit jpeg only files in camera. They *WILL* lose DR.

That's why most of us save in raw. But not all people do and so they lose
information. It's unavoidable.

Exactly my point all along.

But it's irrelevant to those who are content with the image from the
camera as is.

Polaroid did very well with Land cameras. You got what you got and that
was it.

Likewise slide film - very narrow range (a little more than 5 stops).
Yet, many of my best film images were from slide.

On the other hand there are a lot of photographers who are satisfied with
the in-camera result or at least with the narrow range of change they can
do with the in-camera JPG. There's nothing 'wrong' with that if it meets
their needs.

Sure, but claiming they don't lose dynamic range (or only 1 or 2 bits) is
what I objected to.

Fair enough. One can argue the number of bits lost or headroom or
shadow depth til the cows come home.

It just does not matter.

Where JPG does lose DR is when it maps to a colorspace. JPG's created
in
camera are mapped to a colorspace (and most cameras provide at least a
few
choices). Such colorspaces have comparatively narrower DR.

Since many only map Jpeg to Srgb, that is indeed true.

Most DSLR's now save in a variety of user selected RGB spaces.

Can't see much point in non standard jpegs myself. I only use RAW for my
use, and use Srgb when converting to jpeg for the internet, because anything
else is pointless.

As your needs are satisfied you shouldn't care about anyone elses.

Or leave it at default since you'll take care of all of it in raw instead
and how it saved in JPG is moot.

Not moot when I save it as Jpeg in PS, and my camera never does anyway. Not
once in over ten years have I found it necessary to ever save as Jpeg in a
DSLR. And some can now convert files from RAW to Jpeg in camera if I ever
do. :-)

You seem to be deliberately missing the point. If raw is important to
you, then that's your need and choice. But, again, to those satisfied
with the in-camera JPG, that is sufficient to them.


Perhaps if I ever have a need to shoot a huge burst rate/length that is
beyond my RAW capabilities, but can be met with Jpeg I would use it.
Otherwise throwing away up to half the camera's performance holds no appeal
for me.

And again your needs are not those of others.


--
"Civilization is the limitless multiplication of unnecessary necessities."
-Samuel Clemens.

On 2012-07-23 00:39 , Trevor wrote:
"Alan Browne" wrote in message
...
How is an 8 bit file "1-2 stops less" than a 12-13 bit one?

A good JPG doesn't "lose" dynamic range because of the number of bits,

Yes it does. A generic 8 bit gamma corrected image will
not have more than about 11 stops of dynamic range.

But how many camera's can correct gamma to your choice before saving to
Jpeg? And why would you want to stuff around with gamma for every shot
whilst shooting rather than save raw and do it later?
Saving to Jpeg in camera is simply for those who don't know any beter, or
whose camera's can't do anything else IMO.

See my other reply - but the main point is the curve is preset in camera.

Exactly, and a one size fits all, can often mean one size fits nothing.

If it meets the needs of that user, it fits them quite well. A lot of
press photographers shoot JPG only - even for color print. It's less
hassle, smaller files (often need to transmit over spotty comms) and so
on. Speed is of the essence and the images are more that good enough
(and still allow for a lot of tweaking if less so than raw).

--
"Civilization is the limitless multiplication of unnecessary necessities."
-Samuel Clemens.

"Alan Browne" wrote in message
...
2. You don't select the curves in-camera, it's pre-programmed.
Exactly, which *IS* the problem!
No it isn't.

It *is* the problem for those claiming dynamic range is not affected.


. For the raw user it is irrelevant.

Now that much is true at least.


That's why most of us save in raw. But not all people do and so they
lose
information. It's unavoidable.

Exactly my point all along.

But it's irrelevant to those who are content with the image from the
camera as is.

If they are happy yes, not when they claim there is no loss of dynamic range
however.


Polaroid did very well with Land cameras. You got what you got and that
was it.

Yep, and were outsold well and truly by film camera's.


Likewise slide film - very narrow range (a little more than 5 stops).

5 stops, hell what film were you using? But yes slides were definitely
inferior to current digital. That's why I don't use it any more.


Yet, many of my best film images were from slide.

How sad for you. Whilst I have *many* great photo's taken on film of all
sorts (it's the image after all) I sure wish I had todays equipment 40 years
ago! I can only imagine what people will be able to take for granted 40
years from now.


Can't see much point in non standard jpegs myself. I only use RAW for my
use, and use Srgb when converting to jpeg for the internet, because
anything
else is pointless.

As your needs are satisfied you shouldn't care about anyone elses.

I don't, neither do I care for those who make false claims.


Not moot when I save it as Jpeg in PS, and my camera never does anyway.
Not
once in over ten years have I found it necessary to ever save as Jpeg in
a
DSLR. And some can now convert files from RAW to Jpeg in camera if I ever
do. :-)

You seem to be deliberately missing the point.

Nope, simply correcting the change of subject.


If raw is important to
you, then that's your need and choice. But, again, to those satisfied
with the in-camera JPG, that is sufficient to them.

Never said otherwise.


Perhaps if I ever have a need to shoot a huge burst rate/length that is
beyond my RAW capabilities, but can be met with Jpeg I would use it.
Otherwise throwing away up to half the camera's performance holds no
appeal
for me.

And again your needs are not those of others.

Never said otherwise, although many simply do not know any better. Jpegs are
fine for the P&S brigade of course.

Trevor.

"Alan Browne" wrote in message
...
Exactly, and a one size fits all, can often mean one size fits nothing.
If it meets the needs of that user, it fits them quite well.

Fine of course, except I'm always amazed at how bad many peoples photo's
are, and how often they complain, so maybe many are not all that happy at
all. Suits me though, as it gives me more work :-)


A lot of press photographers shoot JPG only - even for color print.

Yes I know, quality is not needed or possible in newspaper printing. I'd
always take RAW+Jpeg at least, because you never know when a photo will be
worth more than tomorrows landfill.


It's less hassle, smaller files (often need to transmit over spotty comms)
and so on. Speed is of the essence and the images are more that good
enough (and still allow for a lot of tweaking if less so than raw).

And thankfully not necessary at all with a modern DSLR, and todays storage
costs. But old habbits die hard for some.

Trevor.

On 2012-07-23 20:17 , Trevor wrote:
"Alan Browne" wrote in message
...
2. You don't select the curves in-camera, it's pre-programmed.
Exactly, which *IS* the problem!
No it isn't.

It *is* the problem for those claiming dynamic range is not affected.

Editing prior posters content to exclude context is plain rude.

With that ...


--
"Civilization is the limitless multiplication of unnecessary necessities."
-Samuel Clemens.

"Alan Browne" wrote in message
...
2. You don't select the curves in-camera, it's pre-programmed.
Exactly, which *IS* the problem!
No it isn't.

It *is* the problem for those claiming dynamic range is not affected.

Editing prior posters content to exclude context is plain rude.

Editing stuff superflous to what you are replying to is curteous. Not doing
so is rude. As is complaining what others choose to do when they probably
think the same of your actions but weren't rude enough or stupid enough to
bitch about it.

Trevor.

On Wed, 25 Jul 2012 11:31:37 +1000, "Trevor" wrote:


"Alan Browne" wrote in message
m...
2. You don't select the curves in-camera, it's pre-programmed.
Exactly, which *IS* the problem!
No it isn't.

It *is* the problem for those claiming dynamic range is not affected.

Editing prior posters content to exclude context is plain rude.

Editing stuff superflous to what you are replying to is curteous. Not doing
so is rude. As is complaining what others choose to do when they probably
think the same of your actions but weren't rude enough or stupid enough to
bitch about it.

It is considered permissible to remove text from the article to which
you are replying if you in some way indicate that you have done so.
e.g.
--- snip ---

[...]

... and similar.

'Editing' someone else's text by deleting some of it without
indicating the fact is commonly used by less scrupulous subscribers to
help them unfairly win arguments. It's a cunning way of lying.

If you delete parts of someone else's text without indicating the
fact, you run the risk of people thinking you are trying to run a
dishonest argument. I have no idea of whether or not that is what you
are trying to do.

As you say, editing stuff to remove unnecessary text from the article
is both a good idea and good nettiquette, but doing it in such a way
that you leave yourself open to suspicions of dishonesty is not.
--

Regards,

Eric Stevens

On 2012-07-24 21:31:54 -0700, Eric Stevens said:

On Wed, 25 Jul 2012 11:31:37 +1000, "Trevor" wrote:


"Alan Browne" wrote in message
...
2. You don't select the curves in-camera, it's pre-programmed.
Exactly, which *IS* the problem!
No it isn't.

It *is* the problem for those claiming dynamic range is not affected.

Editing prior posters content to exclude context is plain rude.

Editing stuff superflous to what you are replying to is curteous. Not doing
so is rude. As is complaining what others choose to do when they probably
think the same of your actions but weren't rude enough or stupid enough to
bitch about it.

It is considered permissible to remove text from the article to which
you are replying if you in some way indicate that you have done so.
e.g.
--- snip ---

[...]

... and similar.

'Editing' someone else's text by deleting some of it without
indicating the fact is commonly used by less scrupulous subscribers to
help them unfairly win arguments. It's a cunning way of lying.

If you delete parts of someone else's text without indicating the
fact, you run the risk of people thinking you are trying to run a
dishonest argument. I have no idea of whether or not that is what you
are trying to do.

As you say, editing stuff to remove unnecessary text from the article
is both a good idea and good nettiquette, but doing it in such a way
that you leave yourself open to suspicions of dishonesty is not.

However, sniping/editing to deliberately change the context of the
dialog is pure gamesmanship.


--
Regards,

Savageduck

"Eric Stevens" wrote in message
...
On Wed, 25 Jul 2012 11:31:37 +1000, "Trevor" wrote:
"Alan Browne" wrote in message
om...
2. You don't select the curves in-camera, it's pre-programmed.
Exactly, which *IS* the problem!
No it isn't.

It *is* the problem for those claiming dynamic range is not affected.

Editing prior posters content to exclude context is plain rude.

Editing stuff superflous to what you are replying to is curteous. Not
doing
so is rude. As is complaining what others choose to do when they probably
think the same of your actions but weren't rude enough or stupid enough to
bitch about it.

It is considered permissible to remove text from the article to which
you are replying if you in some way indicate that you have done so.
e.g.
--- snip ---

[...]

... and similar.

'Editing' someone else's text by deleting some of it without
indicating the fact is commonly used by less scrupulous subscribers to
help them unfairly win arguments. It's a cunning way of lying.

If you delete parts of someone else's text without indicating the
fact, you run the risk of people thinking you are trying to run a
dishonest argument. I have no idea of whether or not that is what you
are trying to do.

As you say, editing stuff to remove unnecessary text from the article
is both a good idea and good nettiquette, but doing it in such a way
that you leave yourself open to suspicions of dishonesty is not.


Well anyone can go back and check if they want to accuse me of being
dishonest. Just bitching because someone doesn't post exactly the same as
they do is another mater entirely.

Trevor.