Re: Photos (Jpg, Png-viewer) --- i don't like it because it launches sluggishly...

J. J. Lodder <nospam@de-ster.demon.nl> wrote:

Antonio Marques <no_email@invalid.invalid> wrote:
 

J. J. Lodder <nospam@de-ster.demon.nl> wrote:

Antonio Marques <no_email@invalid.invalid> wrote:
 

Rich Ulrich <rich.ulrich@comcast.net> wrote:

On Thu, 25 Jul 2024 21:47:38 -0600, Tilde <invalide@invalid.invalid>
wrote:
 

Tony Cooper wrote:

On Thu, 25 Jul 2024 15:19:17 -0700, HenHanna <HenHanna@devnull.tb>

There is a  (Windows) tool called  Photos (Jpg, Png-viewer)  --- i
don't like it because it launches  sluggishly....
 
Esp. in the last 5 days or so, i'm noticing that almost every day
i have to go to Properties to  change it back to
my fav. Jpg, Png-viewer tool
 
because the Windows update (?) is  pushing  Photos  on me.
 
is there a Fix for this???

 
I have thousands of images from .jpgs to .pngs on my computer.  I use
the (free) FastStone Photo Viewer.  It's not only a great image
viewer, but offers many other options from selecting by tagged images
to bulk re-naming.  It's set as my default viewer.
 
https://www.faststone.org/

 
https://www.irfanview.com/

 
I have both Faststone and Irfanview, and I like Faststone better.
 
What I remember last using Irfanview for was when I wanted
to change the default orientation of some pictures that were
usually wrong (downloaded from my off-brand phone).
 
IIRC, Faststone would rotate them okay for PC display by Faststone,
but they would be wrong when uploaded to Face Book.

 
Opening and saving a lossy format like jpg will usually result in... loss
of quality.

 
That's ancient folklore, from the times when 640x480 was a big image.
It may get noticable, but only when you order a huge reduction
in file size,

 
....no, it's the logical and unavoidable result of applying a lossy
encoding, all the more since the jpeg algorithm won't be the exact same
every time, and will throw out slightly different parts of the signal. It
will obviously be worse the lower the resolution is to begin with, but
that's a different issue.

 
You are merely regurgitating theory.

That remark says more about you than about me.

I've pointed out the name of a tool that rotates jpegs losslessly. Because,
you know, reencoding with a lossy codec is lossy. You chose that hill to
die on. OK, you do you.

Have you ever tried to have a look at it?

Such as what, the endless stream of ruined jpegs people keep sharing on the
Web?

Such as the pictures that people 'scan' with $50 scanner and then spend
hours retouching, ending up with an oil painting?

'Everybody' knows that jpeg is not lossless, and therefore -BAD-.
Few people ask themselves: -what is it- that is being 'lost'.
(and is that good or bad)

Fewer people even express a preference for loss while denying it's loss,
but here we've found one!

Years ago on hydrogenaudio some guy said his sister preferred her music
with a low pass filter. It's perfectly legitimate. She didn't try to claim
it was crystal clear.

 
In many cases the loss of so called 'information'
is actually a good thing.
There is 'information' and 'information'. [1]

If you understood the issue, you'd know that 'loss of information' doesn't
mean 'smoothing of data'. Yes, smoothing can be loss, but in the case of
digital codecs actual loss is the difference between the input and the
output, which, more than missing data, has extraneous data, sometimes known
as 'artifacts'.

In reducing RAW data to best quality jpeg,

Nobody even suggested the conversation was about RAW -> jpeg. In fact, the
conversation wasn't even specifically about photos.

what the jpeg keeps in mostly actual, real, image information. [2]
What jpeg 'loses' is mostly noise, such as sensor noise
and quantisation noise.

Not at all. Digital photography noise shows up beautifully in jpegs, and
compounded. It's areas without noise such as skies and lit expanses that
tend to fare the best. Because they're full of light. The sensor isn't
scavenging for photons. There are few abrupt edges or dots, that cause the
most problems to the jpeg transform.

The problems always start with shadows.

Apart from that there is usually a huge amount of redundancy
in an image that encodes each pixel separately,
with for example 14 bits/pixel.
Typical example: the blue sky may take up half the magabytes
of your raw sensor data.
Almost all the 'information' in those bits is redundant,
and most of what isn't redundant is noise.
Reducing the redundancy (and averaging the noise)
doesn't involve any real loss of image quality. [3]

It's telling that your 'example' is of an extreme case. Plus, redundancy of
data is a boon to any compression algorithm, not just lossy ones. You may
be surprised to find that not all your RAW files have the same size.

At any rate, it's utterly irrelevant to the discussion, which was never
about RAW -> jpeg. To be honest I don't even know where you got that from,
the few references to personal photography were mostly about film.

 
Jan
 
[1] Remember that is picture which consists of nothing but noise
contains the most 'information' of all, in the informational sense
of 'information'.
[2] Yes, I know, there are obvious exceptions,
such as when your camera has a stuck pixel.
[3] Extreme example: consider an image that is uniformly medium grey.
Jpeg will reduce the 'information content' in the form of file size
by 99.99999%, to just a few bytes, with no loss at all
of picture content.

Actually, cartoon-like images are where jpeg's flaws show up more easily,
and that's why those are sensibly distributed as PNG. Actual photos or
realistic drawings with gradients is pretty much where anyone will use jpeg
these days. Unless, of course, they don't know what they're doing.

The moral: it isn't as simple as you seem to think it is.
 

I have absolutely no idea what you think I think 'the moral' is, all the
more since I didn't even know there was a moral at stake.

Apparently your moral is that one should rotate one's jpeg photos every day
with a destructive tool, because each time will throw out more unwanted
data.