Picture Size (Pixels & Kb)

Ken Lucke wrote:
In article , Ron Hunter
wrote:

Justin C wrote:
On 2007-01-27, Ken Lucke wrote:
In article . com,
tallmanirl wrote:

Hello everyone,
what is the relationship between the no. of pixels a picture has, it's
width and height and the Kb it takes up, esp. pix on the Web.
Zero.
Gotta say, Ken, you're, at least, concise!

Justin.

but concisely wrong...

Really?

Then can you tell me what the relationship is between how many pixels a
picture has and its height & width, with simply that question asked,
without information on aspect ratio and dpi/ppi?

How about how many KB it takes up from its number of pixels, given the
information in the original question doesn't tell you what kind of file
format, what compression level setting, or the actual picture
information (which is highly critical in the amount of compression that
can be achieved at various "compression settings")?

How about the relationship between the height & width and the number of
KB it takes up, given the information offered lacks any other reference
points?


Oh. sorry, then I must have been wrong in saying that there was zero
relationship between those things.

To say there is zero relationship between those factors certainly is
wrong. But isn't that what I said? Concise, but wrong.

Ken Lucke wrote:
In article A14vh.1335$li4.1050@trndny08, Dave Cohen
wrote:

Toke Eskildsen wrote:
Ron Hunter wrote:

Ken Lucke wrote:
Well, it IS the correct answer for the question asked. There is
NO relationship between those things without many other factors
being specified.

So I can have a gigabyte pixel image on the web with zero bytes
size?
You're making a formal mistake here. While there might be no
relationship, you'll still have to obey the limitations of the chosen
image format.

Unless we get into the whole semi-philosophical debate about the empty
group satisfying any rules (sorry about the bad translation, english is
my second language), then all image formats takes up some bytes.


So your gigapixel image can be compressed to a single colored square,
specified as having the right dimensions and taking up a few bytes,
or it could be stored uncompress and taking up gigabytes. This would
satisfy the NO relationship clause.

Not too realistic, is it?
The problem is that it really _is_ realistic, if you only look at
megapixels. A single-colored square at gigapixel size would likely be
compressed down to a few KB using PNG (or TIFF LZW or GIF or...).

So we can't say anything, unless we have an idea of the nature of the
original image and the expected quality.
For a post about about which we can't say anything, an awful lot of
posters are making the attempt. Why do have to go on and on with posts
of this type. A simple one thread reply of insufficient parameters
specified would have sufficed.
Dave Cohen

Which was what my original "zero" response did - it indicated that
there was no way of establishing any relationship between those things
without further data. Others have blown out the context and tried to
make it some sort of an issue - it's not. It is simply impossible to
answer the original, badly phrased and horribly under-data'd question
as it stood.

It is not that there IS no relationship between the factors, but that
the relationship, although present, depends on other factors. In normal
cases, the more pixels an image has, the larger the file that results,
whether or not the file is compressed, and without regard to compression
method, or subject matter, or file format. There are special cases
where this relationship will not be as expected, but it exists, in the
vast majority of cases, so the relationship is hardly zero, only
difficult to define precisely.

Ken Lucke wrote:
In article , Ron Hunter
wrote:

Ken Lucke wrote:
In article m, Justin
C wrote:

On 2007-01-27, Ken Lucke wrote:
In article . com,
tallmanirl wrote:

Hello everyone,
what is the relationship between the no. of pixels a picture has, it's
width and height and the Kb it takes up, esp. pix on the Web.
Zero.
Gotta say, Ken, you're, at least, concise!


Well, it IS the correct answer for the question asked. There is NO
relationship between those things without many other factors being
specified.

So I can have a gigabyte pixel image on the web with zero bytes size?
Not too realistic, is it?

No, but you can't have a relationship described without all the factors
necessary for that relationship being known. The OP asked a question
that had no more possibility of being accurately ansered than the
question "how much blue do I need?" - which is zero posibibility.
Hence the annswer "zero".

In the vast majority of cases, there does exist a direct relationship to
the number of pixels, and file size, and the width and height of a
picture, when displayed. Other factors intervene, but I don't believe
the OP was interested in special cases, or exceptions, but rather the
way in which these factors affect the relationship.

Ron Hunter wrote:

It is not that there IS no relationship between the factors, but
that the relationship, although present, depends on other factors.

Which means that there is no relationship, when we only look at the
original factors.

You're speaking beyond the stated premise, which is a good thing - it
educates and if you're lucky, you might even hit whatever the original
poster wanted to know, but didn't have the knowledge to formulate.
But it does not make the zero answer wrong.

It might have been more forthcoming though, such as "It depends on a
lot of factors, such as x, y and z".
--
Toke Eskildsen - http://ekot.dk/

In article , Toke
Eskildsen wrote:

Ron Hunter wrote:

It is not that there IS no relationship between the factors, but
that the relationship, although present, depends on other factors.

Which means that there is no relationship, when we only look at the
original factors.

[realizing you are not actually responiding to me in this article]
Which is precisely true... there are MANY things in life that have zero
relationship - until you add one or more additional factors.

You're speaking beyond the stated premise, which is a good thing - it
educates and if you're lucky, you might even hit whatever the original
poster wanted to know, but didn't have the knowledge to formulate.
But it does not make the zero answer wrong.

Thank you. :^)

It might have been more forthcoming though, such as "It depends on a
lot of factors, such as x, y and z".

You are right in that respect, but my point was to attempt to make the
OP actually *think* about his question and WHY that particular answer
was given (and correct).

--
You need only reflect that one of the best ways to get yourself a
reputation as a dangerous citizen these days is to go about repeating
the very phrases which our founding fathers used in the struggle for
independence.
-- Charles A. Beard