dots per inch?

On Sep 19, 7:49 am, bugbear <bugbear.DeleteThis@trim_papermule.co.uk_trim> wrote:

>
> You didn't mention the correct units for sensor resolution
>
> BugBear

Which I would say would be "samples per inch."
 >> Stay informed about: dots per inch? 

On Sep 19, 8:27 am, dj_nme <dj_... RemoveThis @hotmail.com> wrote:
> Don Stauffer in Minnesota wrote:
>
> > On Sep 19, 7:49 am, bugbear <bugbear RemoveThis @trim_papermule.co.uk_trim> wrote:
>
> >>You didn't mention the correct units for sensor resolution
>
> >> BugBear
>
> > Which I would say would be "samples per inch."
>
> How about "sensel pitch", usually stated in nanometres
> (nm)?

I think you meant micrometers ( AKA microns ), 0.000001 meters.
 >> Stay informed about: dots per inch? 

Scott W added these comments in the current discussion du jour
....

> eugene wrote:
>> Someone has asked me to give them an image with 350 dots per
>> inch. Is this the same as 350 pixels per inch?
>>
>
> There will be no doubt another long debate about ppi vs. dpi,
> but yes if they asked for 350 dots per inch it is a very good
> bet that what they really meant was 350 pixels per inch.
>
re: another long debate. I think you may be right here, Scott! As
best I can tell, the term DPI originated in the printing industry
for the earliest of what we now know as half-tone printing, which
is the laying down of a pattern of "dots" in a geometrically
regular pattern at an angle, often 45 degrees. In the days of B &
W newspapers, the number and size of the "dots" laid down roughly
to approximate the shades of gray in a continuous tone photo of
the day allowed the black dots and white news print to fool the
human eye into thinking they were looking at a real photo.

Later, the same idea was applied to color photos.

Much, much later, when computer scanners were invented, the term
DPI now had real meaning, in that it specified how many "dots"
the user wanted the scanner to "sample" or scan per linear inch
across the paper and down its length. Each "dot" resulted in a
pixel in the final image. AFAIK, that definition of "DPI" is
still valid.

I won't go into the debate about DPI for ink jet printers except
to say that the number of "dots" laid down by the printer in
order to "dither" to get the desired colors and density isn't at
all the same sort of definition I used above. Yes, of course the
printer is actually producing N DPI but it seems difficult for
them that aren't mathematicians to understand what is really
happening.

For us normal folk, who just want to get "good" prints from a
given pixel resolution raster graphics image, I agree with you
and most others in that "DPI" really means "PPI". I do not get
any of my images printed by pro shops nor do I dabble in
"printing" my images in publications that use today's version of
half-tone printing, but I suppose there are places that are set
up to calculate the finished print size using "DPI". One can
obviously do the exact same thing using "PPI" but perhaps older
print machines and/or older software may want DPI.

And with that, let the debate begin!

--
HP, aka Jerry
 >> Stay informed about: dots per inch? 

HEMI-Powered wrote:
> Scott W added
>>eugene wrote:
>>
>>>Someone has asked me to give them an image with 350 dots per
>>>inch. Is this the same as 350 pixels per inch?
>>
>>There will be no doubt another long debate about ppi vs. dpi,
>>but yes if they asked for 350 dots per inch it is a very good
>>bet that what they really meant was 350 pixels per inch.

There can only be one interpretation if asked for a digital file. It
would only mean printer dot resolution if the request was for a paper print.

> re: another long debate. I think you may be right here, Scott! As
> best I can tell, the term DPI originated in the printing industry
> for the earliest of what we now know as half-tone printing, which
> is the laying down of a pattern of "dots" in a geometrically
> regular pattern at an angle, often 45 degrees. In the days of B &
> W newspapers, the number and size of the "dots" laid down roughly
> to approximate the shades of gray in a continuous tone photo of
> the day allowed the black dots and white news print to fool the
> human eye into thinking they were looking at a real photo.

Interestingly, half-tone requires a huge resolution in digital files
because it's based on the photographic process of developing through a
screen with a grid of holes... with a film transparency overlaid. The
brighter areas burn each dot down to a smaller dot, the dark areas make
bigger black dots. And it's measured in lpi (lines per inch) just to
make things more confusing... halftones can also be done with lines
where the line simply gets fatter or thinner according to the exposure.
Digital is not good at achieving this because digital can only jump by
full pixels and cannot make gradual size increments like film.

> Later, the same idea was applied to color photos.
>
> Much, much later, when computer scanners were invented, the term
> DPI now had real meaning, in that it specified how many "dots"
> the user wanted the scanner to "sample" or scan per linear inch
> across the paper and down its length. Each "dot" resulted in a
> pixel in the final image. AFAIK, that definition of "DPI" is
> still valid.
>
> I won't go into the debate about DPI for ink jet printers except
> to say that the number of "dots" laid down by the printer in
> order to "dither" to get the desired colors and density isn't at
> all the same sort of definition I used above. Yes, of course the
> printer is actually producing N DPI but it seems difficult for
> them that aren't mathematicians to understand what is really
> happening.
>
> For us normal folk, who just want to get "good" prints from a
> given pixel resolution raster graphics image, I agree with you
> and most others in that "DPI" really means "PPI". I do not get
> any of my images printed by pro shops nor do I dabble in
> "printing" my images in publications that use today's version of
> half-tone printing, but I suppose there are places that are set
> up to calculate the finished print size using "DPI". One can
> obviously do the exact same thing using "PPI" but perhaps older
> print machines and/or older software may want DPI.
>
> And with that, let the debate begin!




--
Paul Furman Photography
http://edgehill.net
Bay Natives Nursery
http://www.baynatives.com
 >> Stay informed about: dots per inch? 

On 2007-09-19 01:33:21 -0700, "eugene" <eugene RemoveThis @home.com.invalid> said:

> Someone has asked me to give them an image with 350 dots per inch. Is
> this the same as 350 pixels per inch?
>
> thanks

For high-end printing (on a conventional press) the resolution was
typically 300-400 dpi at the imagesetter. Exactly equivalent to 300-400
pixels per inch in a TIFF file. My (Canon) 6-color inkjet printer
produces a similar look from an image file whose resolution is as low
as 144 ppi, and in fact higher res than that simply increases file size
(slows down printing) w/o any increase in quality.

YMMV. To test your own set-up, output images at 72, 144 and 300 ppi,
with the highest quality settings and the best paper you intend to use.
My iP6700D does a really nice job with International Paper products
branded "National Geographic." I buy it when it's on sale at the same
price as Epson papers. For uncoated I like Hammermill Great White.
--
Cease then to grieve for your private afflictions, and address
yourselves instead to the safety of the republic
 >> Stay informed about: dots per inch? 

HEMI-Powered wrote:
>
> And with that, let the debate begin!
>

You didn't mention the correct units for sensor resolution

BugBear
 >> Stay informed about: dots per inch? 

Don Stauffer in Minnesota wrote:
> On Sep 19, 7:49 am, bugbear <bugbear.DeleteThis@trim_papermule.co.uk_trim> wrote:
>
>
>>You didn't mention the correct units for sensor resolution
>>
>> BugBear
>
>
> Which I would say would be "samples per inch."

How about "sensel pitch", usually stated in nanometres (nm)?
I think it depends on the manufacturer, as this seems to be rarely
described as a feature of a camera's sensor by most manuafacturers, some
describe it also as a "pixel pitch", it's usually only described with an
overall megapixel (mp) rating and it's inch-size (as in: 2/3") or crop
factor (as in: 1.5x).
 >> Stay informed about: dots per inch? 

"eugene" <eugene.DeleteThis@home.com.invalid> wrote in message
news:upednfyRr6VMQ23bnZ2dnUVZ8qminZ2d@bt.com...
> Someone has asked me to give them an image with 350 dots per inch. Is this
> the same as 350 pixels per inch?
>
> thanks

No. Which program are you using? You should be able to change it to whatever
they want without resizing.

On second thought, maybe they did mean they wanted it 350 pixels wide,. Is
it for web?
 >> Stay informed about: dots per inch? 

In article <Zd9Ii.9124$z_5.8493@nlpi069.nbdc.sbc.com>, paul- DeleteThis @-edgehill.net
says...

>There can only be one interpretation if asked for a digital file. It
>would only mean printer dot resolution if the request was for a paper print.


The first sentence above is correct, dpi can only mean pixels per inch there.
The second is not correct, or at least is quite hasty and rather incomplete.

There can only be one interpretation of dpi as concerning any image file, at
least if from a scanner or a camera, any and all those images we have to work
with. Both terms dpi or ppi are only about paper, because the only place
inches exist is on the paper. Regarding images, all inches are on paper.
Technically, the image file only has pixels.

Certainly the term can never mean ink drops if about image files. There are
no dithered ink drops in image files, which I think was your meaning. But we
do print those image files to our printers too, at perhaps 300 dpi, meaning
pixels per inch on paper. Then after inside the printer, there may be
another definition that comes into play to create the ink drops to do that
job of printing 300 pixels per inch on paper.

This whole notion of sematics about dpi is really dumb. For example, my big
dictionary has 116 definitions for the word "set". First one is to put
something in a particular place, and the last one is stubborn or obstinate.

The analogy is that some of us (I do not mean you Paul, I think your
intentions were good, but I mean others in past days) are obstinate,
insisting words can only have one meaning, specifically insisting the only
possible valid definition is the one they are able to understand. And when
newbies figure out there are two definitions, some of them go crazy inventing
their own rules for the rest us to abide by. Which is nonsense, because in
the real world, image resolution has always been called dpi (meaning pixels
per inch), and always will be called dpi. Just a name, but that is simply
the name that has always been used. So much easier to just accept that it is
used, because it definitely is used.

Saying PPI is fine too, no problem at all with ppi, that is indeed what it
means, but we absolutely must understand it either way, ppi or dpi, because
we absolutely will hear it both ways. It is very much like the word "set"
too, we need to understand the usage context determines the meaning.

I do not argue about which is "correct". Both are correct. Ppi may be more
clear to newbies, use it if you wish, but the name of the term has always
been dpi, and it is fully correct too. Always has been.

I only argue that both terms are used with the same meaning.

The point being that the only correct answer to such questions is that it is
called both dpi and ppi, either one interchangeably. Pick one and use it
yourself, but understanding that both are used is essential, or else we don't
know much, and we will remain confused.

--
Wayne
http://www.scantips.com "A few scanning tips"
 >> Stay informed about: dots per inch? 

jdear64 wrote:
> On Sep 19, 8:27 am, dj_nme <dj_... RemoveThis @hotmail.com> wrote:
>
>>Don Stauffer in Minnesota wrote:
>>
>>
>>>On Sep 19, 7:49 am, bugbear <bugbear RemoveThis @trim_papermule.co.uk_trim> wrote:
>>
>>>>You didn't mention the correct units for sensor resolution
>>
>>>> BugBear
>>
>>>Which I would say would be "samples per inch."
>>
>>How about "sensel pitch", usually stated in nanometres
>>(nm)?
>
>
> I think you meant micrometers ( AKA microns ), 0.000001 meters.

It depends on who wrote the camera review article, they may use nm or
um, depending on whether they wish the pixel/sensel pitch to look large
or small.
 >> Stay informed about: dots per inch? 

sheepdog 2007 added these comments in the current discussion du
jour ...

> For high-end printing (on a conventional press) the resolution
> was typically 300-400 dpi at the imagesetter. Exactly
> equivalent to 300-400 pixels per inch in a TIFF file. My
> (Canon) 6-color inkjet printer produces a similar look from an
> image file whose resolution is as low as 144 ppi, and in fact
> higher res than that simply increases file size (slows down
> printing) w/o any increase in quality.

I have a Canon Pixma 6600 which I find to produce superb images
on all the glossy Canon papers I've tried. Since my primary
subject is cars and other subjects that look "sharper" on glossy
paper, I've not done very much with the semi-glass and matte
papers other than just run a few tests.

It's refreshing to hear that I am not the only one who is happy
with relatively low PPI. I normally create my finished images at
1400 x 1050, no larger than 1600 x 1200 unless I know in advance
I will be printing for max quality. So, for a borderless 8.5 x 11
print, the PPI is pretty dismal - under 130. Yes, or course I can
see aliasing, especially on the near-horizontal or vertical
character lines and chrome moldings on my cars. But, since I
print so rarely and view from a distance where the aliasing is
not bothersome to me (it would be clearly unacceptable to most
people based on what I've read on this NG), I am happy as the
proverbial clam.

> YMMV. To test your own set-up, output images at 72, 144 and
> 300 ppi, with the highest quality settings and the best paper
> you intend to use. My iP6700D does a really nice job with
> International Paper products branded "National Geographic." I
> buy it when it's on sale at the same price as Epson papers.
> For uncoated I like Hammermill Great White.

I already "rang in" with my views on the differences between
"DPI" and "PPI", so please excuse my hyjacking this thread. I
changed the subject so that no one would read my post thinking it
was a continuation of the DPI/PPI debate.

My question for you, sheepdog, is what exactly is "National
Geographic" paper? I can see that you said it is manufactured by
International Paper but I cannot recall seeing NG as a brand of
their paper. Is it ultra-glossy, glossy, semi-gloss, matte, or
what? Most importantly, how is the price/sheet compared to
Canon's brand and how would you compare quality?

When I am printing test prints or photos where I don't need the
higher cost of glossy paper, I have also found Hammermill to be
about the finest non-photo paper, although it is pricier than
normal "copier" paper.

One other question: Do you use the High or Standard print quality
setting in the driver? Or, have you created your own custom
settings? My 6600 just drinks ink like it was water, as do most
high-quality photo printers, so I also tested both driver
settings on the same paper with the same car. Besides the paper
being wetter and the print time much higher, I could scarcely see
any difference so I have gone to Normal as my "normal" (couldn't
resist the pun!) setting. Besides, it seems to make no sense to
waste so much expensive ink on a print coming from an image with
only 1/2-2/3 the PPI that ordinarily be needed for a "good"
print.

Your comments would be appreciated, and again, I apologize to
those who may be upset that I hyjacked this thread, it's just
that so much of what you talk about I find to be equally true.

Have a great day!

--
HP, aka Jerry
 >> Stay informed about: dots per inch? 

just bob added these comments in the current discussion du jour
....

>> Someone has asked me to give them an image with 350 dots per
>> inch. Is this the same as 350 pixels per inch?
>>
> No. Which program are you using? You should be able to change
> it to whatever they want without resizing.
>
> On second thought, maybe they did mean they wanted it 350
> pixels wide,. Is it for web?
>
I agree that DPI and PPI generally have two completely different
meanings, but in the photo editors I've used, I can change the PPI
(some actually do call it DPI) with or without resampling. Quite
some time ago, when I was using Adobe PhotoShop 5 LE that came with
my old Microtek scanner, changing the PPI was the only way to alter
the print size.

Since I do not use custom print shops, I've not run into the
problem really being discussed here, but I do occasionally change
the print size to something in inches that has meaning to me by
first unchecking the "resample" box. One reason I do that is to
save the picture in the correct aspect ratio for an 8 x 10 or 8.5 x
11 but leave enough space for the narrower aspect ratio of a 4 x 6.
Thus, it helps me to know how much to crop to get an image that
will (almost) exactly fill the printed sheet.

--
HP, aka Jerry
 >> Stay informed about: dots per inch? 

"HEMI-Powered" <none DeleteThis @none.en> wrote in message
news:Xns99B1401C82EC5ReplyScoreID@140.99.99.130...
> just bob added these comments in the current discussion du jour
> ...
>
>>> Someone has asked me to give them an image with 350 dots per
>>> inch. Is this the same as 350 pixels per inch?
>>>
>> No. Which program are you using? You should be able to change
>> it to whatever they want without resizing.
>>
>> On second thought, maybe they did mean they wanted it 350
>> pixels wide,. Is it for web?
>>
> I agree that DPI and PPI generally have two completely different
> meanings, but in the photo editors I've used, I can change the PPI
> (some actually do call it DPI) with or without resampling. Quite
> some time ago, when I was using Adobe PhotoShop 5 LE that came with
> my old Microtek scanner, changing the PPI was the only way to alter
> the print size.
>
> Since I do not use custom print shops, I've not run into the
> problem really being discussed here, but I do occasionally change
> the print size to something in inches that has meaning to me by
> first unchecking the "resample" box. One reason I do that is to
> save the picture in the correct aspect ratio for an 8 x 10 or 8.5 x
> 11 but leave enough space for the narrower aspect ratio of a 4 x 6.
> Thus, it helps me to know how much to crop to get an image that
> will (almost) exactly fill the printed sheet.

Right , I resize without resampling the same way: I usually print on 13x19"
paper and change the dimensions to 12x8" so I get a half inch border.

If I take a Canon 1DM3 file that is 3888 x 2592 pixels and Bridge with ACR
converted the RAW file at 72 dpi, the Photoshop resize box says the
dimensions are 54x36 inches at 72 "pixels per inch". When I uncheck Resample
Image and change the dimensions to 12x8"", the pixels per inch changes to
216. In this case 216 is my DPI.

Now what has me confused is the new Getty photo service which sells a photo
to anyone up to "72dpi up to 500KB" for $49. How can they specify dpi
without specifying the dimensions? How/why can they make a general
statement that it will be about 500KB? I can't see why they wouldn't just
say, "you're file will be up 400 pixels max dimension and it's saved at JPEG
quality 5", or something similar, which, IMO, seems like a better way to
control the quality of the file you sell and it's how I've dealt with web
advertisers.
 >> Stay informed about: dots per inch?