How to connect to a wireless device (a video camera's memory card)from windows 10?

In article , Commander Kinsey
wrote:

But selling a camera that makes an image with 10 million pixels in it,
simply
upsampled from a 2.5MP sensor, is cheating.

it's not doing that

My last Fuji admitted to it.

it did no such thing.

It did, it was called interpolation.

it didn't, and yet another thing you don't understand.

They claimed they had some fancy **** using
hexagonally shaped arrays of pixels on the CCD.

but not 2.5mp

It was sold as a 6MP camera with a 3MP CCD. What it gave out was about as
good as 4MP from a real camera.

moving the goalposts once again.

your 10mp camera, a fuji finepix s1000fd, was sold as a 10mp camera:

In article , Commander Kinsey
wrote:
Just try it yourself. Or view mine. This is the number 6 from the bottom of
my wall clock. Download it, open it in photoshop, and view it at 100%. Now
view it at 50% and notice it's sharper. This was taken with a Fuji Finepix
S1000fd which claims 10MP. Clearly when you view it at 2.5MP (50%) you see a
better quality image. It isn't capable of producing 10MP.

http://www.fujifilm.com/support/digi...tions/s/finepi
x_s1000fd/

You might aswell buy a 2.5MP camera.

no

Yes, because you can always upsample in photoshop.

that has nothing to do with the camera

Yes it does.

nope. what you do in photoshop is entirely separate.

If a 10MP camera gives out something you could create by
upsampling an image from a 5MP camera, then the 10MP camera is no
better than the 5MP camera.

except it's not doing that.

In article , Commander Kinsey
wrote:


And a hard disk is used on a computer, so they should use binary
notation, just like they do with RAM.

some operating systems do.

In article , Sjouke Burry
wrote:


Besides,they can always give you the number of color pixels, three
times the full pixel density.

it doesn't work that way

"nospam" wrote in message
...
In article , Commander Kinsey
wrote:


If I take a photo with the camera set to 2.5MP, then enlarge that in
photoshop to 10MP, I don't have a 10MP image.

yes you do.

It won't be as good
as a photo from a 10MP camera.

irrelevant.

But selling a camera that makes an image with 10 million pixels in it,
simply
upsampled from a 2.5MP sensor, is cheating.

it's not doing that

You might aswell buy a 2.5MP camera.

no

I agree with Commander Kinsey. If the sensor has a native resolution of 2.5
MP, then no matter how much you upscale it, you won't be adding any more
detail. A 100x100 pixel image, upscaled to 10 MP, will still only contain
the same detail as the original 100x100 image. Upscaling doesn't magically
add any detail.

If a camera outputs a 10 MP JPG but from a 2.5 MP sensor, you have a 2.5 MP
image (in terms of the amount of detail) upscaled to 10 MP. That is *not*
the same as an image from a 10 MP sensor.

Do some tests to demonstrate to yourself that the two images are not the
same: that the true 10 MP image contains extra detail that the
2.5-upscaled-to-10 image does not have.

On 1/7/2019 5:22 AM, NY wrote:
"nospam" wrote in message
...
In article , Commander Kinsey
wrote:


If I take a photo with the camera set to 2.5MP, then enlarge that in
photoshop to 10MP, I don't have a 10MP image.

yes you do.

It won't be as good
as a photo from a 10MP camera.

irrelevant.

But selling a camera that makes an image with 10 million pixels in
it, simply
upsampled from a 2.5MP sensor, is cheating.

it's not doing that

You might aswell buy a 2.5MP camera.

no

I agree with Commander Kinsey. If the sensor has a native resolution of
2.5 MP, then no matter how much you upscale it, you won't be adding any
more detail. A 100x100 pixel image, upscaled to 10 MP, will still only
contain the same detail as the original 100x100 image. Upscaling doesn't
magically add any detail.

If a camera outputs a 10 MP JPG but from a 2.5 MP sensor, you have a 2.5
MP image (in terms of the amount of detail) upscaled to 10 MP. That is
*not* the same as an image from a 10 MP sensor.

Do some tests to demonstrate to yourself that the two images are not the
same: that the true 10 MP image contains extra detail that the
2.5-upscaled-to-10 image does not have.
Following up on this, if you display that image from a 10mp camera on a
screen that has a 2.5 MP resolution, it will appear the same as an image
from a 2.5 MP sensor.

Where you see the difference is when you zoom both images to 100 X, the
10MP image will appear better on that 2.5MP screen than the 2.5 MP image
on the same screen.

While theoretically I was aware of this for years, when it became
obvious was when we were house hunting. I would take pictures of the
inside and outside on the house we were looking at. When I returned
home, I would answer my questions about the house by zooming in the the
part of the house that was in question. ie are there street lights in
front of the house. What kind of faucet is in the bath room. With the
image from the 10MP sensor, you could zoom in on that part and answer
your question. With the 2.5MP image you could not get a picture clear
enough to identify the object you were looking at.

In another example from the same house hunting project, I was not aware
of the street lights when we visited the house. By zooming the pictures
from the that had the street in the background, I could find the street
lights, with the images from the 2,5 mp image I could not find them due
to the pixelation of the image.

--
2018: The year we learn to play the great game of Euchre

In article , NY
wrote:

If I take a photo with the camera set to 2.5MP, then enlarge that in
photoshop to 10MP, I don't have a 10MP image.

yes you do.

It won't be as good
as a photo from a 10MP camera.

irrelevant.

But selling a camera that makes an image with 10 million pixels in it,
simply upsampled from a 2.5MP sensor, is cheating.

it's not doing that

You might aswell buy a 2.5MP camera.

no

I agree with Commander Kinsey.

he's wrong.

If the sensor has a native resolution of 2.5
MP, then no matter how much you upscale it, you won't be adding any more
detail.

that's true, however this is not about detail, it's about number of
pixels.

the amount of detail in an image has nothing to do with how many pixels
it has.

a 2.5mp image upsampled to 10 mp is now a 10 mp image.
similarly, a 10 mp image downsized to 2.5 mp is now a 2.5 mp image.

take an image and blur it, enough that most (or even all) detail is
lost. it still has the same number of pixels, but with less detail.

examples:
https://blog.udemy.com/wp-content/uploads/2014/05/box-blur.png
https://img.kelbymediagroup.com/plan...014/10/BandA_0
3.jpg

he is also claiming that the camera is cheating. that is false. it's
not cheating.

On Mon, 07 Jan 2019 16:25:41 -0000, Wolf K wrote:

On 2019-01-06 18:42, Commander Kinsey wrote:
[...]
My last Fuji admitted to [cheating]. They claimed they had some fancy ****
using hexagonally shaped arrays of pixels on the CCD.
[...]

Hexagonal packing is a method of getting more dots into the same area,
so you get a higher physical resolution. So with the same lenses, a
hex-array chip of the same size as a square-array chip will capture more
detail (which you may perceive as a sharper image, depending on circs.)

Maybe so, but Fuji were then adding more fake pixels to that, computed by the camera's internal chip. They actually admitted it was a 3MP sensor giving out 6MP interpolated. To be honest it did somehow manage about 4MP worth, but nothing like 6.

On Mon, 07 Jan 2019 16:15:38 -0000, Wolf K wrote:

On 2019-01-06 18:26, Commander Kinsey wrote:
On Sun, 06 Jan 2019 23:23:54 -0000, nospam wrote:

In article , Commander Kinsey
wrote:

the fuji *is* a 10mp camera.

It gives out 10 million pixels

therefore it's a 10mp camera.

but they aren't unique.

spatially, they are.

their value is irrelevant.

If I take a photo with the camera set to 2.5MP, then enlarge that in
photoshop to 10MP, I don't have a 10MP image.

yes you do.

It won't be as good
as a photo from a 10MP camera.

irrelevant.

But selling a camera that makes an image with 10 million pixels in it,
simply upsampled from a 2.5MP sensor, is cheating. You might aswell buy
a 2.5MP camera.

You've got it backwards. When you set the camera to a lower resolution,
it downsamples the 10MP to 2.5MP (or whatever the setting is.)

I've got nothing backwards. The 10MP output is NOT better than the 2.5MP output. Therefore the sensor and/or lens just can't handle 10MP. They're lying through their teeth because perhaps it's a 10MP sensor, but either a really **** one with interference, or the lens just doesn't make a sharp enough image to define dots that small.

In article , Wolf K
wrote:


Besides,they can always give you the number of color pixels, three
times the full pixel density.

I think you mean the dots. A pixel is 3 dots, each specified by one
byte, hence 3x8-- 24BPP (bits per pixel) colour depth. Ditto for a
monitor: each pixel is 3 dots.

nope.

a pixel, short for picture element, is a spatial element of an image.

a pixel can have one or more components (what you're incorrectly
calling dots), depending whether it's monochrome, rgb, rgba, cmyk,
hexachrome or something else, and is not necessarily one byte per
component either. 2 bytes per component (16bpc/64bpp) or floating point
values are both common.

a typical colour display usually has 3 components per pixel (rgb), but
that's not always the case.

an oled display has more and smaller green than red and blue:
http://m.xslightings.com/Content/upl...22175202495997
1.jpg
https://www.oled-info.com/files/styl...mages/Samsung-
pentile-matrix-subpixel-closeup.jpg

quattron adds yellow for a supposedly wider gamut. it was also a
commercial failure and you probably never heard of it.
https://upload.wikimedia.org/wikiped...-color_pixel_s
tructure.jpg

and of course, a monochrome display only has 1 component per pixel.

Dots per inch do matter: two monitors of
different sizes but with the same resolution will have different size
dots. Hence at the same close-in viewing distance, the larger one will
look less sharp.

that part is true, although with hi-dpi displays, where individual
pixels are smaller than what the eye can resolve, you'd need to be
*really* close to see a difference, possibly even needing
magnification.

In article , Commander Kinsey
wrote:


But selling a camera that makes an image with 10 million pixels in it,
simply upsampled from a 2.5MP sensor, is cheating. You might aswell buy
a 2.5MP camera.

You've got it backwards. When you set the camera to a lower resolution,
it downsamples the 10MP to 2.5MP (or whatever the setting is.)

I've got nothing backwards.

true. you have it entirely wrong.

The 10MP output is NOT better than the 2.5MP
output.

it is if you don't **** it up.

Therefore the sensor and/or lens just can't handle 10MP. They're
lying through their teeth because perhaps it's a 10MP sensor, but either a
really **** one with interference, or the lens just doesn't make a sharp
enough image to define dots that small.

nonsense.

"Wolf K" wrote in message
...
On 2019-01-06 18:42, Commander Kinsey wrote:
[...]
My last Fuji admitted to [cheating]. They claimed they had some fancy
**** using hexagonally shaped arrays of pixels on the CCD.
[...]

Hexagonal packing is a method of getting more dots into the same area, so
you get a higher physical resolution. So with the same lenses, a hex-array
chip of the same size as a square-array chip will capture more detail
(which you may perceive as a sharper image, depending on circs.)

I've always wondered: with hexagonal packing, where every alternate row of
pixels is offset half a pixel-width horizontally from the row above, how do
they map that offset structure to the rectangular grid of pixels that is
assumed for any display or for any image-processing algorithm? Is there a
matrix transform which will interpolate the value that a pixel would have if
its sensor was directly underneath another one rather than being offset
horizontally and vertically by half a pixel?

In article , NY
wrote:

Hexagonal packing is a method of getting more dots into the same area, so
you get a higher physical resolution. So with the same lenses, a hex-array
chip of the same size as a square-array chip will capture more detail
(which you may perceive as a sharper image, depending on circs.)

I've always wondered: with hexagonal packing, where every alternate row of
pixels is offset half a pixel-width horizontally from the row above, how do
they map that offset structure to the rectangular grid of pixels that is
assumed for any display or for any image-processing algorithm? Is there a
matrix transform which will interpolate the value that a pixel would have if
its sensor was directly underneath another one rather than being offset
horizontally and vertically by half a pixel?

yes.

On Mon, 07 Jan 2019 16:13:10 -0000, Wolf K wrote:

On 2019-01-06 19:26, Sjouke Burry wrote:
On 6-1-2019 22:33, Mike wrote:
On 1/6/2019 9:34 AM, nospam wrote:
In article , Commander Kinsey
wrote:
[...]
For example, take a photograph with 10MP camera. Use photoshop to
change
that to 2.5MP (by resizing the image to 50%). Now resize it back to
10MP.
If the camera was really 10MP, the adjusted image should be crap
compared to
the original, but it's identical. Therefore the original image only
contained 2.5MP. Blocks of 4 pixels were the same and gave no extra
information.

nonsense. it is in no way identical.


Besides,they can always give you the number of color pixels, three
times the full pixel density.

I think you mean the dots. A pixel is 3 dots, each specified by one
byte, hence 3x8-- 24BPP (bits per pixel) colour depth. Ditto for a
monitor: each pixel is 3 dots. Dots per inch do matter: two monitors of
different sizes but with the same resolution will have different size
dots. Hence at the same close-in viewing distance, the larger one will
look less sharp.

No need to check dots per inch. That can be calculated from screen size and resolution.

On Mon, 07 Jan 2019 17:25:51 -0000, NY wrote:

"Wolf K" wrote in message
...
On 2019-01-06 18:42, Commander Kinsey wrote:
[...]
My last Fuji admitted to [cheating]. They claimed they had some fancy
**** using hexagonally shaped arrays of pixels on the CCD.
[...]

Hexagonal packing is a method of getting more dots into the same area, so
you get a higher physical resolution. So with the same lenses, a hex-array
chip of the same size as a square-array chip will capture more detail
(which you may perceive as a sharper image, depending on circs.)

I've always wondered: with hexagonal packing, where every alternate row of
pixels is offset half a pixel-width horizontally from the row above, how do
they map that offset structure to the rectangular grid of pixels that is
assumed for any display or for any image-processing algorithm? Is there a
matrix transform which will interpolate the value that a pixel would have if
its sensor was directly underneath another one rather than being offset
horizontally and vertically by half a pixel?

I'd guess it probably messes it up. Remember trying to record interlaced video from a TV onto your computer and getting every other line slightly squint when things moved rapidly in the image? There's a name for it which I can't remember, which some video editors can reduce, but they never get rid of it entirely.

On Mon, 07 Jan 2019 17:18:36 -0000, nospam wrote:

In article , Commander Kinsey
wrote:


But selling a camera that makes an image with 10 million pixels in it,
simply upsampled from a 2.5MP sensor, is cheating. You might aswell buy
a 2.5MP camera.

You've got it backwards. When you set the camera to a lower resolution,
it downsamples the 10MP to 2.5MP (or whatever the setting is.)

I've got nothing backwards.

true. you have it entirely wrong.

The 10MP output is NOT better than the 2.5MP
output.

it is if you don't **** it up.

Therefore the sensor and/or lens just can't handle 10MP. They're
lying through their teeth because perhaps it's a 10MP sensor, but either a
really **** one with interference, or the lens just doesn't make a sharp
enough image to define dots that small.

nonsense.

Thankyou once again ROD SPEED for your complete and utter lack of input.