O/T: Nibbling on an Apple

In article , Eric Stevens
wrote:

if you print it, then the ppi is determined by the number of pixels /
size of print. if you don't print it, then there is no ppi because
there are no inches.

What does print have to do with anything. Stop twisting.

i'm not twisting a thing. ppi is only relevant when printing. if you're
not printing, it's meaningless.

Actually the quoted figure of 72ppi is relevant to the older Apple
displays. (It may still be relevant but I don't know that). The
Microsoft standard was 1/3 greater - 96dpi - for the usual Microsoft
reasons.

it's was never relevant.

Many older CRTs used to have a pitch of 72dpi but towards the end they
began creeping up. My present LCD has 92dpi.

that's very low.

htc one is 468 ppi
retina iphone is 326 ppi
nikon d7000 rear lcd is 267 ppi
15" retina macbook pro is 220 ppi.
microsoft surface pro is 208 ppi
11" non-retina macbook air is 135 ppi.
27" lcd display is 111 ppi

On Tue, 13 Aug 2013 18:43:37 +0200, Sandman wrote:

In article ,
nospam wrote:

In article , PeterN
wrote:

if you print it, then the ppi is determined by the number of pixels /
size of print. if you don't print it, then there is no ppi because
there are no inches.

What does print have to do with anything. Stop twisting.

i'm not twisting a thing. ppi is only relevant when printing. if you're
not printing, it's meaningless.

Ehm, that's the other way around. "PPI" is for displays and monitors.
How many pixels are shown per inch on a display. That is irrelevant to
printing. Your monitor showing 350 PPI doesn't mean anything when you
print the image

The thing which nobody has specifically stated is that the actual
pixels/inch used by the monitor are fixed. They are built into the
hardware. So if the native display pitch of your monitor is 96 ppi
that's all it will ever display. It will ignore instructions to
display at (say) 72dpi simply because it can't do that. That's why if
the file doesn't change the image size displayed on the screen remains
constant as it is dependent only on the number of pixels to be
displayed.

It is often possible to set up your monitor to display at a different
resolution from the native pitch but it will be still using its native
(say) 96 ppi. What it will be doing is resampling your image on the
fly with a probably cheap and nasty (but fast) algorithm. This will
result in the displayed image being inferior to the one you would see
if it was displayed at the native resolution.

"DPI" means "Dots per inch" and is only relevant to printing. How many
separate ink dots it can put per inch of media.

It used to mean that but not for some time. Print colors used to be
created by putting down dots of ink in 'cells'. See
http://www.rags-int-inc.com/PhotoTec.../DotMatrix.gif
which displays cells 4x4 and 2x2 in size. The number of cells roughly
corresponds to pixels in the original image. But then as in the
example, printers arrived with the capability to vary the size of the
dots. This enabled the printer to vary the size of the cells. Matters
got even worse when printers became able to drop dots almost anywhere
without adhering to any particular dot pitch. By this time cells lost
any direct connection with pixels and could instead be regarded as
vari-colored patches which could be laid down to form the image. Color
boundaries were no longer tied to cell boundaries and could traverse
cells.

Dots per inch increasingly became a meaningless figure whose principal
use was for competitive bragging between printer manufacturers.

I have an Epson 3800 (which is now obsolescent). This has native cell
sizes of 360/inch, 720/inch, 1440/inch and a claimed 2880/inch.
According to the ppi selected when the print driver is set up one gets
nothing but large droplets, a mixture of large and small droplets, or
only small droplets. What's more, even for uniform colours, the
droplets do not appear to be deposited in in a consistent pattern from
one cell to the next.

I get the impression that the selected print ppi is a conversion
factor which will enable you to calculate the size of the print image
from the number of pixels and a guide as to the detail quality one may
expect. But it's not in any way an indication of the DPI of what is
laid down by the printer or the number of dots one may expect to be
deposited.

A 100x100 image will be 1 inch x 1 inch on a 100 PPI screen and print as
1 inch x 1 inch when printing as a 100 DPI image on a printer.

http://en.wikipedia.org/wiki/Pixel_density

http://en.wikipedia.org/wiki/Dots_per_inch

A digital image can record a DPI value which sometimes is referred to as
both DPI and PPI since neither really applies - it's just a command for
if and when the digital image should be representaed in the physical
world.
--

Regards,

Eric Stevens

In article , Eric Stevens
wrote:

The thing which nobody has specifically stated is that the actual
pixels/inch used by the monitor are fixed.

i said it yesterday or the day before.

They are built into the
hardware. So if the native display pitch of your monitor is 96 ppi
that's all it will ever display. It will ignore instructions to
display at (say) 72dpi simply because it can't do that. That's why if
the file doesn't change the image size displayed on the screen remains
constant as it is dependent only on the number of pixels to be
displayed.

which is why some computers are no longer using pixels for imaging.
they're using points, which may be more than one pixel. this is because
individual pixels can no longer be seen.

It is often possible to set up your monitor to display at a different
resolution from the native pitch but it will be still using its native
(say) 96 ppi. What it will be doing is resampling your image on the
fly with a probably cheap and nasty (but fast) algorithm. This will
result in the displayed image being inferior to the one you would see
if it was displayed at the native resolution.

not necessarily.

try a retina display sometime.

I get the impression that the selected print ppi is a conversion
factor which will enable you to calculate the size of the print image
from the number of pixels and a guide as to the detail quality one may
expect.

that's what it's for.

But it's not in any way an indication of the DPI of what is
laid down by the printer or the number of dots one may expect to be
deposited.

it's not supposed to be. dpi is a function of the printer.

In article 2013081321351088124-savageduck1@REMOVESPAMmecom,
Savageduck wrote:

Savageduck really does think he knows more than the software in the
camera.

I'll let you know when I find that camera that can figure out my intentions.

Your D300S auto-focuses, does it not? Is that not your intention, to
shoot in-focus pictures? And if you don't want it to autofocus, don't
you turn it off? Can you turn it off? Isn't it actually on most of the
time?

You also use Auto-ISO when shooting some motor sport shots, that way
software adapts the ISO to match your intentions, does it not?

All of these are software features YOU decide when to use, and you have
them on when and if they match your intentions. So if there is a
function that can automatically detect blinking people when you're
taking a quick family portrait and thus save you time to review the
images and call back all the people because two people were blinking,
then that's your intention, and the software can help you because you
can't detect that.

I pretty much always shoot in manual mode, and I want complete control
of my camera, but that doesn't mean I can see a use for automatic
functionality.

I rarely manually focus, though.

--
Sandman[.net]

On Wed, 14 Aug 2013 00:37:15 -0400, nospam
wrote:

In article , Eric Stevens
wrote:

Exposure, focus etc are the result of a physical measurement. Whether
or not an image is crap is not determined by measurement but often by
subjective judgement. I don't see how you can automate that.

if you want over- or underexposed photos or out of focus photos, don't
use something that filters them.

I do get over or underexposed photographs when I push the limits. I
don't always throw them away but try to salvage them in
post-processing.

you still can do that.

meanwhile, the rest of the world likes their photos properly exposed
and in focus and won't have a problem with something that flags photos
that aren't.

Then they will miss some damn good photographs.

not necessarily, and most people aren't interested in pushing the
limits. they want well exposed and in focus images.

That's fine, as long as you don't claim that those who can make images
out of photographs that are less than properly focussed or exposed are
at fault for not wanting to use software that rejects photographs that
are less than well exposed or focussed.

it's not judging a photo on composition. it just filters out the
mistakes. everyone has some of those.

They are not always mistakes.

then don't delete them.

learn how to properly use such a tool and when to use it, rather than
dismiss it outright without even considering how it could be useful.

By all accounts the tool deletes what it regards as less than perfect
photographs before the photographer has a chance form an opinion of
their own.

I don't follow. You are letting the camera make decisions about
whether or not an image is junk, yet you say the user is in control. I
would have though that meant the user was making decisions.

the user decides when and where to use a given feature and how it's
configured.

modern autofocus systems have several options, including single focus
versus continuous, how many focus points to use, whether the camera
tracks the subject or the user sets the focus point, etc.

But the resulting shots may not be all perfect but, even then, some of
them may still be highly desirable.

nothing is perfect.

But some are desirable.

all autofocus needs to do is do a better job than humans do, which it
definitely does in most situations.

there are edge cases where it might not be appropriate, and then you
just turn it off.

face detection with autofocus is another example. normally, when
someone takes photos of people, they want to focus on their face, not
the trees behind them or whatever. the camera can detect the face and
use that as a focus target. if you don't want that and prefer the trees
to be in focus, don't enable the feature. very simple.

it's a tool. learn how to use it and it works very well. if you don't
learn how to use it then it might not do what you want. no surprise
there. don't blame the camera when you don't know what you're doing. if
you tell it to do something, it will do it. if you don't want it to do
that, don't tell it to do it.

The decision not to tell it is mine.

yep.
--

Regards,

Eric Stevens

On Tue, 13 Aug 2013 21:02:23 -0700, Savageduck
wrote:

On 2013-08-13 20:49:10 -0700, Eric Stevens said:

On Tue, 13 Aug 2013 09:53:31 -0700, Savageduck
wrote:

On 2013-08-13 09:28:00 -0700, Sandman said:

In article ,
nospam wrote:

they don't want to
relinquish control.

I thought automatic cars were quite popular in the USA more so then the UK,
not sure about other countries.

automatic transmissions are popular but that's not the same.

people fear self-driving cars. they're legal in 3 states so far, but
you can be sure the average person isn't going to want to get into one
any time soon, despite the fact they are *much* safer than when a human
drives.

Same goes for airplanes. Computers can now take off, fly and land
airplanes 100% without human intervention, and most of the latest
airplane accidents have been due to pilots taking control over the plane
where the computer probably would have handled the situation a lot
better.

But I don't think a single person in the world would jump on an airplane
where the pilot seats were empty.

Then there are the flying cultures, such as in Korea's Asiana Airlines
where they are so dependent on computer assisted glide slope landings
that they can't handle visual approaches that are second nature to
pilots trained in the West. That is what happened with the recent
Asiana crash at SFO, and the 1999 Korean Airways crash on take-off at
London Stansted Airport.
http://en.wikipedia.org/wiki/Asiana_Airlines_Flight_214

As indicated in the Wikipedia article there seems to have been some
confusion amongst the crew over the setting and operation of the
autothrottle. This is consistent with the events of the crash.

There appears to be a serious cockpit culture issue with both Asiana &
Korean Airlines. The cockpit communications between flight crew and
familiarity with visual approach procedures was decidedly lacking.
Two US pilots who had flown for Asiana after they retired, one from
United and one from Delta, both reported this problem of the Korean
pilots dependance on automated systems. The ex-United pilot reported
that one Korean Asiana couldn't execute a normal visual approach into
LAX. He told the American pilot "I don't need to know that. We don't do
that." They had to go around and the American captain had to make the
visual approach and landing.

Not good.

It sounds a bit like some of the philosophies recently advocated in
this news group for the taking of photographs.
--

Regards,

Eric Stevens

On Wed, 14 Aug 2013 01:12:46 -0400, nospam
wrote:

In article , Eric Stevens
wrote:

The thing which nobody has specifically stated is that the actual
pixels/inch used by the monitor are fixed.

i said it yesterday or the day before.

Sorry: I missed it.

They are built into the
hardware. So if the native display pitch of your monitor is 96 ppi
that's all it will ever display. It will ignore instructions to
display at (say) 72dpi simply because it can't do that. That's why if
the file doesn't change the image size displayed on the screen remains
constant as it is dependent only on the number of pixels to be
displayed.

which is why some computers are no longer using pixels for imaging.
they're using points, which may be more than one pixel. this is because
individual pixels can no longer be seen.

Then why have smaller pixels?

It is often possible to set up your monitor to display at a different
resolution from the native pitch but it will be still using its native
(say) 96 ppi. What it will be doing is resampling your image on the
fly with a probably cheap and nasty (but fast) algorithm. This will
result in the displayed image being inferior to the one you would see
if it was displayed at the native resolution.

not necessarily.

try a retina display sometime.

I bet it's possible to resample an image in a fashion which would make
it appear to its disadvantage on a Retina display. Nevertheless I
accept that in most cases most people would be hard put to notice the
difference.

I get the impression that the selected print ppi is a conversion
factor which will enable you to calculate the size of the print image
from the number of pixels and a guide as to the detail quality one may
expect.

that's what it's for.

But it's not in any way an indication of the DPI of what is
laid down by the printer or the number of dots one may expect to be
deposited.

it's not supposed to be. dpi is a function of the printer.

That's what I have said. It's also a function of how the printer
driver is set up. I've just said that too. :-)
--

Regards,

Eric Stevens

On Wed, 14 Aug 2013 00:37:20 -0400, nospam
wrote:

In article , Eric Stevens
wrote:

if you print it, then the ppi is determined by the number of pixels /
size of print. if you don't print it, then there is no ppi because
there are no inches.

What does print have to do with anything. Stop twisting.

i'm not twisting a thing. ppi is only relevant when printing. if you're
not printing, it's meaningless.

Actually the quoted figure of 72ppi is relevant to the older Apple
displays. (It may still be relevant but I don't know that). The
Microsoft standard was 1/3 greater - 96dpi - for the usual Microsoft
reasons.

it's was never relevant.

Just coincidence - huh?

Many older CRTs used to have a pitch of 72dpi but towards the end they
began creeping up. My present LCD has 92dpi.

that's very low.

htc one is 468 ppi
retina iphone is 326 ppi
nikon d7000 rear lcd is 267 ppi
15" retina macbook pro is 220 ppi.
microsoft surface pro is 208 ppi
11" non-retina macbook air is 135 ppi.
27" lcd display is 111 ppi

So mine is not doing too badly for a four year old 24" display.
--

Regards,

Eric Stevens

In article ,
nospam wrote:

i'm not twisting a thing. ppi is only relevant when printing. if you're
not printing, it's meaningless.

Actually the quoted figure of 72ppi is relevant to the older Apple
displays. (It may still be relevant but I don't know that). The
Microsoft standard was 1/3 greater - 96dpi - for the usual Microsoft
reasons.

it's was never relevant.

It is true that the original Macintosh had a 72 ppi screen, but it isn't
true that the "Microsoft standard" was 96 ppi, because Microsoft sold no
displays (as far as I know.

A 17" screen with the resolution of 1280x1024 has a ppi of 96, but back
when the 72/96 ppi myth started, there were no such high resolution
screen, and when they were common, the original Macinosh was just a
memory and the screens were as common for Macs as for PCs.

You need to go down to a 13" screen at 1024x768 to get 96ppi again, but
13" desktop screens weren't very common (15" was the common step). 13"
was more for laptops, and again, when people had laptops to start such a
myth, the original Mac was again a memory.

Many older CRTs used to have a pitch of 72dpi but towards the end they
began creeping up. My present LCD has 92dpi.

that's very low.

Indeed, 92ppi is quite low. That's a 20" at 1600x900 or 24" at
1920x1080. I can't think of any other resolution that would match and be
current.

I once wrote a PPI calculator that I just remember:

http://sandman.net/misc/ppi

That's to calculate the PPI of your screen, or any screen.




--
Sandman[.net]

In article ,
Whisky-dave wrote:

most people are.

I don;lt

The number one reason I can never take any posts from drunk Dave
seriously.

"I don;lt"

Exactly

"sarcasm in lieu of a coherent argument "

I wasn't sarcastic. Nor was I part in the argument you had with nospam.
I just jumped in with a pejorative against your abysmal spelling, which
means that reading your posts is more like decoding hieroglyphs than
reading a coherent line of thinking.

I can only assume that before posting to usenet you down a bottle of
Whiskey and then type with your face.


--
Sandman[.net]