Can we improve screenshot DPI

Bram van den Heuvel wrote:
Given news wrote:

He wants better screenshots. Isn't that
enough reason? The problem seems to be
that he doesn't understand that he's not
taking a picture of something, but is rather
just copying the pixels displayed onscreen.

I appreciate that you're helpful since some of the others don't even seem
to understand the question where they think printing is involved or where
they think it matters what you're screenshotting.

The question is just one of what determines resolution of any given
screenshot.

Specifically, how do I increase the number of pixels copied off the screen?

Someone asked "what" I'm taking a screenshot of, where I can't imagine that
his question has any technical merit since it shouldn't matter what you're
taking a screenshot of since the screen is displaying it - so you can't do
better than the screen, right?

Someone else also mentioned printing, which again has no bearing on the
question since the question is only about how to increase the resolution of
the captured screenshot.

Let's say I have a current screenshot resolution of X.
And let's say I want to double that resolution to 2X.

How can a person do that?

They're not going to double their resolution by what they're looking at.
They're not going to double their resolution by printing it.

How can you double the resolution of a screenshot?

Do you change a software driver?
Do you double your screen size?
Do you double your memory?

What determines the resolution of any given screenshot on your own screen?

You can make a virtual screen (a pan and scan screen),
where the size of the desktop the OS uses, is larger than the screen.

When you move the mouse around, where it hits the edge of the screen,
the viewport moves so you can see a different part of the screen.

This is an example of taking a screenshot of a 16000 pixel high
(160") Firefox window. The original picture is 1440x15900 pixels.
Since this site will not accept pictures that big, it is scaled
by a factor of 14 (that's why it is fuzzy). The intention is not
to show a clear picture, merely to show how much info can be
captured in one screenshot.

https://s17.postimg.org/r2ixgy3lb/linux.gif

The one in Windows was done with a printer driver. I can get
108" high pages with that. It's possible that 108" times 300DPI
is the coordinate space limit at some step in the process.

https://s18.postimg.org/kyjya0gex/sample3.gif

To make a pan and scan setup in Windows, at one time you
needed to "enable unavailable resolutions" in a control panel.
And when you selected one of those, you could get a larger
virtual surface. But it would be subject to the video card
coordinate limits. Early cards were 2048 or so. Later cards
might have supported slightly larger values (the value doesn't
have to correspond to an output crossbar value, like 4K Displayport
or something).

When the Linux setup supported 16000, I was surprised. It implies
the addressing inside the video card goes that high. And the Linux
setup didn't "like" what I was doing. I ran a vanilla Xorg with no
DE, and it still didn't behave properly. I had to take my screenshot
with GIMP, as XWD wouldn't work right. And that's a bad sign.

*******

The software has an idea of "what an inch is". If you're using
Photoshop, you might see a scale around the edges. The scale
can be in pixels. Or it can be in inches.

Your monitor, in the EDID, might declare it is 96x96 DPI. And
when the software is doing the math, it uses that scale factor.

You would need a means to interfere with that declaration, to
improve the resolution on screen objects. If you told the OS
(whatever OS you're using), that the monitor is 192 DPI (a Hi-DPI
display), then the Firefox display would end up twice as wide
on a virtual display. In my example above, I can capture
160" @ 96DPI, or I could capture 80" @ 192DPI. Firefox would think
it was displaying 80" of content, and would be driving sufficient
pixels to (what it thinks) is an inch of screen.

There is the possibility of improved resolution by doing that.

But these ideas are not general purpose, and not a "daily driver"
configuration. They're suitable for a bar bet, and that's about it.

I did something like this 20 years ago, in Unix, using XMX or something
and a second X server. I couldn't see the screen. I launched programs
with an XYWH on the command line, to control their position. And I
took pictures of the screen I could not see, with XWD. Try driving
a screen some time, which you cannot see :-) At least in the experiment
this year, I could see the 16000 pixel high screen, 900 pixels at a time.
But the Xorg server really didn't like that, and I could tell sooner
or later, I would pay a price for using that as a 24/7 solution.

*******

In Linux xorg.conf, this has to be added. Of course, a vanilla system
install has no xorg.conf, and I used aticonfig to snag one. (It will
dump one for you.) Then, you edit, and add this to make the display
larger. I don't know if Xorg would start if this was larger,
so you have to experiment.

SubSection "Display"
Virtual 2048 2048
EndSubSection

Then, this is done while the screen is running, to make a huge desktop.
I didn't make any attempt to make a "square surface" here. Just wide
enough for the Firefox window, and as long as it would tolerate.

# Panning on a 1440x16000 desktop while displaying
# 1440x900 mode on an output called CRT1:

xrandr --fb 1440x16000 --output CRT1 --mode 1440x900 --panning 0x16000

*******

And there are people doing this sort of thing on Windows.

https://web.archive.org/web/20151220...ic.php?t=21972

DALNonStandardModesBCD1

Paul

On 8/23/2017 11:28 AM, nospam wrote:
In article , Keith Nuttle
wrote:

Some basic about pictures: The quality of the images in any picture is
determined by the lens of the device used to capture a picture. The
ability of the lens to focus the image on the file/sensor determines the
absolute quality of the picture. This is fixed an there is no way to
improve it!

not true.

As I said you need to get a good book on Photograph and the optics.
There is not way to improve the quality of the image that is produced by
the optics focus and the sensor.

If you are thinking pinhole camera, the projected images still follows
the same rules as through the lens system. The difference is in the
lens the light ray is bent as it passes through the lens, the pinhole
does not bend the light ray.

The quality of the recording system whether a digital sensor or chemical
film limits the qualty of the picture. The smaller the grain in the
emulsion, or the number of pixel in the sensor that better the
reproduction of the original scene.


--
2017: The year we lean to play the great game of Euchre

Mayana,
I'm thinking why we make our VB6 manifests DPI-aware in that 96DPI is the
default setting for text size on the desktop, and changing that to say 120DPI
will affect the DPI of screenshots.

--
Garry

Free usenet access at http://www.eternal-september.org
Classic VB Users Regroup!
comp.lang.basic.visual.misc
microsoft.public.vb.general.discussion

"Keith Nuttle" wrote in message
news
On 8/23/2017 11:28 AM, nospam wrote:
In article , Keith Nuttle
wrote:

Some basic about pictures: The quality of the images in any picture is
determined by the lens of the device used to capture a picture. The
ability of the lens to focus the image on the file/sensor determines the
absolute quality of the picture. This is fixed an there is no way to
improve it!

not true.

As I said you need to get a good book on Photograph and the optics. There
is not way to improve the quality of the image that is produced by the
optics focus and the sensor.

If you are thinking pinhole camera, the projected images still follows the
same rules as through the lens system. The difference is in the lens the
light ray is bent as it passes through the lens, the pinhole does not bend
the light ray.

The quality of the recording system whether a digital sensor or chemical
film limits the qualty of the picture. The smaller the grain in the
emulsion, or the number of pixel in the sensor that better the
reproduction of the original scene.


I think all the components in the photographic process have the potential to
be the rate-limiting step.

Take two extremes:

- there is no point if having a superb lens, free of any aberrations such as
astigmatism, chromatic aberration, etc if the sensor is only 800x600: the
sensor resolution will be the rate-limiting step

- there is no point having a 10,000x10,000 pixel sensor if the lens is made
from the bottom of a milk bottle and displays chromatic aberration,
astigmatism, severe barrel or pincushion distortion, flare or general lack
of sharpness

You say "there is not [sic] way to improve the quality of the image that is
produced by the optics focus and the sensor". True, but you can improve the
*apparent* sharpness by edge-enhancement of the digital image, and you can
correct for geometric distortion (barrel/pincushion) in the lens with
software such as PT Lens, at the expense of a little loss of sharpness. You
can also correct for intentional parallelogram distortion in an image with
parallelogram correction in Paint Shop Pro and Photoshop: I've used this
many times if I've had to take a photo of a reflective subject (a brass
plaque or a picture under glass) and low light/lack of tripod has meant that
flash is the only way to get the pictu in order to prevent seeing the
flash gun and/or room surroundings reflected, shoot at an angle and correct
in software, making sure you take a head-on reference photo as well to
correct for the change in aspect ratio than parallelogram correction causes.
Yes, you will lose sharpness, but that may be the lesser of two evils.

On 23/08/2017 15:39, Savageduck wrote:
On Aug 23, 2017, David B. wrote
(in article ):

On 23/08/2017 14:57, Savageduck wrote:
On Aug 23, 2017, Mayayana wrote
(in article ):

wrote

So I will ask once again; what is your purpose for these
higher resolution screenshots, why do you need the higher resolution?

He wants better screenshots. Isn't that
enough reason? The problem seems to be
that he doesn't understand that he's not
taking a picture of something, but is rather
just copying the pixels displayed onscreen.

Obviously the starting point is to have a higher resolution display, and the
something he is trying to capture should not be postage stamp size.

Can I change this format?

https://youtu.be/uRucAkHZIp4

Your question should be; When will I remember to have my iPhone correctly
oriented when shooting video?

I didn't forget ....... I didn't *KNOW*!!!

Thanks for the pointer, Savageduck.

--
“Men occasionally stumble over the truth, but most of them pick
themselves up and hurry off as if nothing ever happened.� (Winston S.
Churchill)

http://imgur.com/a/5pb8b

On Wed, 23 Aug 2017 13:30:44 -0400, Keith Nuttle
wrote:

--
2017: The year we lean to play the great game of Euchre

This is way off topic, but I've been wondering this for many months.
Why do you want to lean to play Euchre instead of learn to play
Euchre?

On 23/08/2017 15:05, nospam wrote:
In article , David B.
wrote:

So I will ask once again; what is your purpose for these
higher resolution screenshots, why do you need the higher resolution?

He wants better screenshots. Isn't that
enough reason? The problem seems to be
that he doesn't understand that he's not
taking a picture of something, but is rather
just copying the pixels displayed onscreen.

Obviously the starting point is to have a higher resolution display, and the
something he is trying to capture should not be postage stamp size.

Can I change this format?

https://youtu.be/uRucAkHZIp4

wtf does that have to do with screenshots?

Nothing. I apologise for posting in the wrong place.

Unintentional .... but I DID learn that there's a right and wrong way to
use an iPhone for taking video.

--
“Men occasionally stumble over the truth, but most of them pick
themselves up and hurry off as if nothing ever happened.� (Winston S.
Churchill)

http://imgur.com/a/5pb8b

On Wed, 23 Aug 2017 11:25:13 -0400, Keith Nuttle
wrote:

I always loved the crime movie where they would take a picture at a half
mile, and blow up the license plate on the car and read the license
number. It is not going to happen in real life. Even my 26MP camera has
limits to what can be blown up and still be readable.

I read an interesting article awhile back on the narrow topic of
resolving license plate information when the data simply doesn't exist.
I'll try to paraphrase.

The data set for US license plates is finite and knowable, so step 1 is
to put all of that info into a database. Next, given a blurry or
pixelated digital picture of a license plate, make an initial assumption
about the State that issued the plate. Then run through a series of
iterative guesses as to what the characters and numbers might be, each
time comparing a blurry or pixelated copy of the test image to the real
image. The two images that most closely align will have the highest
probability of being right.

Or something like that. Anyway, it was cool because they said it
attempts to overcome the very issue you highlighted above. They also did
it with partially obscured images, but that has its own challenges.

On 8/23/2017 1:49 PM, Pat wrote:
On Wed, 23 Aug 2017 13:30:44 -0400, Keith Nuttle
wrote:

--
2017: The year we lean to play the great game of Euchre

This is way off topic, but I've been wondering this for many months.
Why do you want to lean to play Euchre instead of learn to play
Euchre?

In Euchre the right and left bowers (The jacks in the suits same color)
are higher ( trumps) than the King, Queen and Ace.


http://www.bicyclecards.com/how-to-play/euchre/

I will let you figure out the rest of the game
--
2017: The year we lean to play t( trumps)eat game of Euchre

On 8/23/2017 1:49 PM, Pat wrote:
On Wed, 23 Aug 2017 13:30:44 -0400, Keith Nuttle
wrote:

--
2017: The year we lean to play the great game of Euchre

This is way off topic, but I've been wondering this for many months.
Why do you want to lean to play Euchre instead of learn to play
Euchre?

Did you ever hear of a spell checker. It can make some quite interesting
changes to what you intend to write.

--
2017: The year we lean to play the great game of Euchre

In article , Keith Nuttle
wrote:

Some basic about pictures: The quality of the images in any picture is
determined by the lens of the device used to capture a picture. The
ability of the lens to focus the image on the file/sensor determines the
absolute quality of the picture. This is fixed an there is no way to
improve it!

not true.

As I said you need to get a good book on Photograph and the optics.
There is not way to improve the quality of the image that is produced by
the optics focus and the sensor.

yes there is.

for example:
http://www.focusmagic.com
Focus Magic uses advanced forensic strength deconvolution technology
to literally "undo" blur. It can repair both out-of-focus blur and
motion blur (camera shake) in an image. It is the only software that
can significantly recover lost detail from blurry images.

https://helpx.adobe.com/photoshop/us...ke-induced-blu
rring.html
Photoshop features an intelligent mechanism to automatically reduce
image blurring caused by camera motion. If necessary, you can adjust
advanced settings to further sharpen the image. The Shake Reduction
filter in the Filter Sharpen menu can reduce blurring resulting
from several types of camera motion; including linear motion,
arc-shaped motion, rotational motion, and zigzag motion.

https://en.wikipedia.org/wiki/Computational_photography
Computational photography or computational imaging refers to digital
image capture and processing techniques that use digital computation
instead of optical processes. Computational photography can improve
the capabilities of a camera, or introduce features that were not
possible at all with film based photography, or reduce the cost or
size of camera elements. Examples of computational photography
include in-camera computation of digital panoramas,
high-dynamic-range images, and light field cameras. Light field
cameras use novel optical elements to capture three dimensional scene
information which can then be used to produce 3D images, enhanced
depth-of-field, and selective de-focusing (or "post focus"). Enhanced
depth-of-field reduces the need for mechanical focusing systems. All
of these features use computational imaging techniques.

If you are thinking pinhole camera, the projected images still follows
the same rules as through the lens system. The difference is in the
lens the light ray is bent as it passes through the lens, the pinhole
does not bend the light ray.

yes it does, called diffraction.

The quality of the recording system whether a digital sensor or chemical
film limits the qualty of the picture. The smaller the grain in the
emulsion, or the number of pixel in the sensor that better the
reproduction of the original scene.

not necessarily.

more pixels in the same size sensor means individual pixels are
smaller, which means they have more noise.

everything is a tradeoff.

In article , GS wrote:

Mayana,
I'm thinking why we make our VB6 manifests DPI-aware in that 96DPI is the
default setting for text size on the desktop, and changing that to say 120DPI
will affect the DPI of screenshots.

no it won't.

it just changes a tag in the image, which is almost always ignored.

In article , David B.
wrote:

Can I change this format?

https://youtu.be/uRucAkHZIp4

Your question should be; When will I remember to have my iPhone correctly
oriented when shooting video?

I didn't forget ....... I didn't *KNOW*!!!

fully explained he
https://www.youtube.com/watch?v=Bt9zSfinwFA

On 23/08/2017 21:47, nospam wrote:
In article , David B.
wrote:

Can I change this format?

https://youtu.be/uRucAkHZIp4

Your question should be; When will I remember to have my iPhone correctly
oriented when shooting video?

I didn't forget ....... I didn't *KNOW*!!!

fully explained he
https://www.youtube.com/watch?v=Bt9zSfinwFA

Thanks - I know *NOW*!

Can you help me change the video format of the swans I recorded? I
believe it can be done .... after the event, as it were.

--
David B.

In article , Wolf K
wrote:

Your question should be; When will I remember to have my iPhone correctly
oriented when shooting video?

I didn't forget ....... I didn't *KNOW*!!!

fully explained he
https://www.youtube.com/watch?v=Bt9zSfinwFA

There's another problem.

When oriented horizontally, with lens to the right or to the left. One
may produce an upside down video when played back on a computer/TV.
There seems to be no consistency. Same problem with horizontal photos.

there is no problem because of the orientation tag in the photo or
video.

if you're seeing it upside-down, then your software is ignoring the tag
which means your software is at fault.